{"as_of":"2026-08-23T19:14:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:7b9911b1210e52d9c93d80faff5d61dee06c2df3dad7ad5ff3324e5b97c9e418","coverage":[{"denominator":62,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":62,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-29T10:18:54.610927Z","state":"measured"},{"denominator":62,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":62,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-23T06:30:58.430688+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2605.27946/citation-record","integrity":"/paper/2605.27946/integrity","json":"/paper/2605.27946/citation-record.json","paper":"/paper/2605.27946"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Uncertainty, neuromodulation, and attention.Neuron, 46(4):681–692, 2005","venue":null,"work_id":null,"year":2005},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:aa17dd1f4875e7c7c749b033b478ad4f1ff1d429c38345b08bd1020f5bcac48d","observation_id":"2295d298-cf58-4b07-b4d7-78c5400c7be1","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Decoupled greedy learning of cnns","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:15c648dd0f54753502880710945b11adeeb0c7ff9aeb493a3c4c4290925dd6fd","observation_id":"111a84dc-8a78-4f66-917c-30e50739dbf0","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1901.09049","last_updated":"2019-02-21T17:15:18Z","snapshot_observed_at":"2026-08-14T17:25:45.886549Z","submitted_at":"2019-01-25T19:07:36Z","title":"Biologically inspired alternatives to backpropagation through time for learning in recurrent neural nets","version":2},"cited_work":{"arxiv_id":"1901.09049","doi":null,"metadata_source":"pith","pith_arxiv_id":"1901.09049","snapshot_observed_at":"2026-06-29T10:23:18.201455Z","title":"Biologically inspired alternatives to backpropagation through time for learning in recurrent neural nets","venue":"cs.NE","work_id":"0e5315fc-9920-4950-bbee-19b1cce2acde","year":2019},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"cited_paper":"/paper/1901.09049","citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:88e31b3372028c7edd54205631d4e57a939c8baf8c44f7da5176a7e1b39455ad","observation_id":"ba691be0-2b74-4dde-884f-055d7994099e","resolution":{"observed_at":"2026-06-29T10:23:18.202655Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"A non-asymptotic analysis of non-parametric temporal-difference learning.Advances in Neural Information Processing Systems, 35:7599–7613, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:e5e209ae4d6d8442e47ec1c8c58c3f4add5d9104dfb04867be9ec0f864cc0e39","observation_id":"d5a93c3e-ccc3-4457-9e32-ba4757fb331a","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"A finite time analysis of temporal difference learning with linear function approximation","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:9d929e5bed4d4526d100d1f0c3f63e30c1e75598602e701ad77803d54d671e5e","observation_id":"ec23af91-6141-4ca4-af30-2a4fdcc6edc8","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"A finite time analysis of temporal difference learning with linear function approximation","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:ad6ddf2fda94f416204ef49a250b0e434d20f3ca36586422bfd37f15a6f35762","observation_id":"e48c7263-8c38-41b5-9aa0-aa0e1e8422fa","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Neural temporal-difference learning converges to global optima.Advances in Neural Information Processing Systems, 32, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:dbf94c77de1e514c047b51444ef8f36d7e5178abf3838b788e45e65f72e55730","observation_id":"224fd486-6f8d-4012-8d14-e502c06bde83","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Finite-time analysis of natural actor-critic for pomdps.SIAM Journal on Mathematics of Data Science, 6(4):869–896, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:d447ec4a78d95fc7a64e1569b958f3445f75add9106ab03bd5d04c43480ff277","observation_id":"4e666946-017d-4aa9-9fae-6d91cb052e49","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"On the statistical benefits of temporal difference learning","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:5557148659246801932139d203d3780fa2fe5477bc3578e0f7219801cc31b940","observation_id":"9931cf42-4d03-4e54-9cfa-a5e3afdca4e1","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Exploring the use of synthetic gradients for distributed deep learning across cloud and edge resources","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:062fc1dee845f07ded8ff70ae0d8ebe00ee5593808a16c7f5768e5ea1edad4e4","observation_id":"1fb09a15-45b7-4546-9818-1a8d434710b8","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"The surprising efficiency of temporal difference learning for rare event prediction.Advances in Neural Information Processing Systems, 37:81257–81286, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:8f86138abef8a9d70145654b7d67b72580a9159678698ee08f3d2569e2ebd5d3","observation_id":"f7b35fdb-5994-4763-9d5d-50df6f2be04e","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Learning phrase representations using rnn encoder–decoder for statistical machine translation","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:91a6e619c652599770925fa691f120d3960f27a0e53933177c8260f9b66aceb1","observation_id":"8b7880dc-d119-4a97-acb9-d861d1fa3887","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Sobolev training for neural networks.Advances in neural information processing systems, 30, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:228a142eb7bbedabe4bf5b9ed7f8e91b66d03648aa0a29a666b9cd231c4fd124","observation_id":"afe1443e-695b-405d-9242-39e0056c173f","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Understanding synthetic gradients and decoupled neural interfaces","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:ce77cd3a0a8af62e797b39693eecf52496bd70e44175efa3cc90a7f90fad575d","observation_id":"f56a4204-7da2-49c1-ab75-9c4426512cf8","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"MIT press, 2005","venue":null,"work_id":null,"year":2005},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:05310df48ca419d7047df268fc3ed78b914131945f0b48cb1812221e345a6818","observation_id":"d7478585-a2d9-4298-9ede-015c1326e09f","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Vectorized instructive signals in cortical dendrites.Nature, 652(8112):1254–1263, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:2922a5758f727a85eaa795f6caeac8ad51312f52ece142c9c718276b43f2f24a","observation_id":"7169e20a-e175-4cd3-90a5-64e7adce1fbb","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Optimality of lstd and its relation to mc","venue":null,"work_id":null,"year":2007},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:d4541ab63f07989332ebc4c1b877532f5e61ac42182fde91992d466668abd20b","observation_id":"f0e1280f-8018-458e-ab89-789ea681b09b","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Deep residual learning for image recognition","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:ab094fbeffb07d7bd26c875188b77353bc8e3b4b61f94caaf7dd9d64a871880f","observation_id":"3526aca3-e792-44f9-b3bf-6353b1bd83df","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2212.13345","last_updated":"2022-12-27T02:54:46Z","snapshot_observed_at":"2026-08-16T16:05:54.047188Z","submitted_at":"2022-12-27T02:54:46Z","title":"The Forward-Forward Algorithm: Some Preliminary Investigations","version":1},"cited_work":{"arxiv_id":"2212.13345","doi":"10.48550/arxiv.2212.13345","metadata_source":"pith","pith_arxiv_id":"2212.13345","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"The forward-forward algorithm: Some preliminary investigations","venue":"cs.LG","work_id":"ca71c444-2040-489b-88f2-7fa8dc57b27a","year":2022},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"cited_paper":"/paper/2212.13345","citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:7b9c065a6e0b8ca21c4351648beeba85c5e5933f0dccc83ac1d38c4a27faf4eb","observation_id":"0c2302f2-dc68-4ade-849d-12d2aa0979b9","resolution":{"observed_at":"2026-06-29T10:23:18.205839Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-05-24T17:53:31.686766+00:00","source":"crossref_status_cache"},{"observed_at":"2026-05-24T17:53:31.686766+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Empirical bayes transductive meta-learning with synthetic gradients","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:5302ee21f4946cef7e9f08d4620aace6c41d0f4043e3ba6b4f2008733395bb30","observation_id":"e189f67a-dc0e-4d41-b83f-2ac89e1477e1","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Decoupled parallel backpropagation with convergence guarantee","venue":null,"work_id":null,"year":2098},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:c90908b187a35a46cf8409147ffa097897e40cf4621c8a38793db91029cef627","observation_id":"12f4cd14-ce90-47b9-9600-807d630b9437","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Decoupled neural interfaces using synthetic gradients","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:cd12576fbbdbb29c00b33b221abfe9472898a43ffa6a7bb9779e5033430908f8","observation_id":"daa78dc1-c1df-4430-be02-fe3650e10223","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Ada-gp: Accelerating dnn training by adaptive gradient prediction","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:8c3c8e00cf5ef960ea85c58815b1f2ede03cfa93023343169fb3479a15d9cf59","observation_id":"87896304-72a0-4135-a546-9e9689d16775","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Synaptic plasticity dynamics for deep continuous local learning (decolle).Frontiers in Neuroscience, 14:424, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:25566e946f32591055013d58bec986ddc7e9045c0b404d265c553acd8b2cb025","observation_id":"ef9d59d0-897e-4caa-939a-bfb7409732ae","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"The fusiform face area: a module in human extrastriate cortex specialized for face perception.Journal of neuroscience, 17(11):4302–4311, 1997","venue":null,"work_id":null,"year":1997},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:7a9c12781c12d457a216252d399be6d97b5aa24d89e7279b16acb144b5a87164","observation_id":"edbda0ca-cc08-4cb3-82ff-60d4df9f78f5","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Transformers are rnns: Fast autoregressive transformers with linear attention","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:5df17849cfb09760c6227927274fc89733115a0c231b0706d150c82059b761fb","observation_id":"9dc2d7b2-e829-4a1a-bbc7-1c5fd78cd982","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Bias-variance error bounds for temporal difference updates","venue":null,"work_id":null,"year":2000},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:df11237495e79e4ce88f08a2b05b3dffa9628c6185d410bef1051acc5f66ae2a","observation_id":"c80e0679-a23a-4bfc-8dc0-ecc226b5cb3e","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Is temporal difference learning optimal? an instance-dependent analysis.SIAM Journal on Mathematics of Data Science, 3(4):1013–1040, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:3ebb46f4907de78a14b9bcaf75e662c6e0ef9fc5e1273d3cac41057a41bf61de","observation_id":"b8fb8abf-30b4-4773-8677-bd7bbb35cd9f","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Presynaptic release probability influences the locus of long-term potentiation.Nature, 360(6399):70–73, 1992","venue":null,"work_id":null,"year":1992},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:ad10dd438453efd58fcb36fc67762c328146a0208f00e1007dd5ea8c12931696","observation_id":"b253d734-17c4-4038-bc90-95d487991696","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Finite-sample analysis of lstd","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:f5a6265b6a3c743b7345aa7d875ebf07203de9fc1d4a7d587ab51bb1edf04304","observation_id":"4d0972ee-8e2f-49c3-8ef5-a66ec171886c","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Deep learning.nature, 521(7553):436–444, 2015","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:ebb99a9750d9bfca90224b9fb481d2e9f5515985ce72055d1d461aeebf29843b","observation_id":"52eac269-ad1f-4950-b949-2574cd652bff","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Backpropagation applied to handwritten zip code recognition","venue":null,"work_id":null,"year":1989},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:28f1b3e9b554b420f6443cdc533e247c4ce78dd48521aab48abd7b122ca46a91","observation_id":"69075038-675e-4eca-a601-f7a80e6edbd8","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Gradient-based learning applied to document recognition.Proceedings of the IEEE, 86(11):2278–2324, 2002","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:c18b8f715e684ddb57ff004d1463f4abf8467c58922a52153e53fc04537fa911","observation_id":"56d4ebe0-5c53-4561-b1ab-e294359d4cd9","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Difference target propagation","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:9d24e31121b3719a933b15b576fa48575010c9372c825c0c555bf8f61522ba43","observation_id":"02d24d12-e622-4765-b2ac-44ee208aafe6","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"High-probability sample complexities for policy evaluation with linear function approximation.IEEE transactions on information theory, 70(8):5969–5999, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:ef4d40db6db9f19227c4fda44cc5217f026f6aad76f8130cf58d5fcf6b33f800","observation_id":"a6ceb875-20b3-4e71-876e-dbc730ed0832","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Accelerated and instance-optimal policy evalua- tion with linear function approximation.SIAM Journal on Mathematics of Data Science, 5(1):174– 200, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:8b251ecd6d74aa5f1f8d25be7e31917606a0b6e8beae5b268999217e1299c87c","observation_id":"6004b075-cadb-4e14-8e30-033595e970f1","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Random synap- tic feedback weights support error backpropagation for deep learning.Nature communications, 7(1):13276, 2016","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:98676e9ca7e7bfbba9cc7635400c56bf253d9e137eb062e2e81311bdeb203392","observation_id":"5c37916d-e5c9-466e-9f6d-e76fd837018f","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Taylor expansion of the accumulated rounding error.BIT, 16(2):146–160, June 1976","venue":null,"work_id":null,"year":1976},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:e18d448b137fc7ae3432d3393dd2d953dd642cc710f54863adb78c3c03be255b","observation_id":"722611e9-8ed6-4555-b2a6-0ae618aad423","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Context-dependent computation by recurrent dynamics in prefrontal cortex.nature, 503(7474):78–84, 2013","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:3e941457b75fa2ee997175809f7585e7ec6c2b56aa00ef8e6889495d12f600fa","observation_id":"853482b6-d20a-4aba-877a-eadb237a6e37","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Synthetic gradient methods with virtual forward-backward networks","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:47f7f9254a7d01383fe9adb8899cc95f2a5ec29b02061b61c2bbcdde9088735a","observation_id":"9459f545-5977-40c9-94ed-1cafb34fff53","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Popgym: Benchmarking partially observable reinforcement learning","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:61792383d2136160fc51a46b3d32336f17a58de3b49e0b7910af6bb0a4ddc02f","observation_id":"9b2acd08-8afb-46b0-95d8-4d2789d0d7b6","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Feed-forward on-edge fine- tuning using static synthetic gradient modules","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:e08741b7f970ff513723afda8d1bf1804e90b90b58525aaca15204d7ab8d6b19","observation_id":"d73e9a98-a8c6-4c33-aa6a-0804916849c4","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"A mesoscale connectome of the mouse brain","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:164d8636c300ff858b05c340c09e1540802fadba021722114e8c1a6390c30b73","observation_id":"70a96700-f120-4854-9431-7c0f985269d4","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Climbing fibers encode a temporal-difference prediction error during cerebellar learning in mice.Nature neuroscience, 18(12):1798–1803, 2015","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:d713e599114604af39aa3f638a7f83c36a2838dfd7f2b1800d180ac777b324ac","observation_id":"5449f1a3-2199-4d0a-a756-30fd8fa60571","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Neural variability and sampling-based probabilistic representations in the visual cortex.Neuron, 92(2):530–543, 2016","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:8793d994f36c16ba79ebdc7897ec62606733e20f33b3588d3d1148d7db990e04","observation_id":"6370902e-987a-4e60-be1c-ea5753e0de1d","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Cortico-cerebellar networks as decoupling neural interfaces.Advances in neural information processing systems, 34:7745–7759, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:214f05e6776f330d970b9e44f8f0a2df63bcb0fcc6a3b1c997dbbe0b68626fda","observation_id":"e651badd-0d44-476f-baf9-4362a2067b95","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"BP($\\mathbf{\\lambda}$): Online learning via synthetic gradients.Transactions on Machine Learning Research, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:97bfe20c3787a8b57e0e26fda91223d8c108a77f3ad375e55958814bc5ba3fbb","observation_id":"2ee06a01-1032-4b44-80b5-4c507934a407","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Learning representations by back-propagating errors.nature, 323(6088):533–536, 1986","venue":null,"work_id":null,"year":1986},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:e7300d9df3d7a75d4a21d973dfa32ddbfc74c5b8c01049294aacf3f5af884696","observation_id":"aba9870a-a945-4f08-99c5-e4e03fd81389","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Equilibrium propagation: Bridging the gap between energy- based models and backpropagation.Frontiers in computational neuroscience, 11:24, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:a5723186b6278528fe8e70fd1a7525a9da273ab8412478d7086af55a2642d8da","observation_id":"98ac0273-18cb-4856-b22a-6457b5459ce5","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Long short-term memory.Neural Comput, 9(8):1735– 1780, 1997","venue":null,"work_id":null,"year":1997},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:abe92e778055033d20c4f214c4f6299f0543b81024e205cfb0bbdc484448cb14","observation_id":"5b76064f-4f8c-4625-976f-609ab8d4b25a","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"A neural substrate of prediction and reward","venue":null,"work_id":null,"year":1997},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:86abf1a84ef6d35011fda31f082aeecbb9b41ee7da28716c5e8e0ee88992482b","observation_id":"c72bddaa-30c5-43d1-8b03-252badc0cd69","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Learning by directional gradient descent","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:5502eefd79937c521aa5fb9e77d5a67a6582a0efa5ff23ee09f084eae02b8d8d","observation_id":"742dfcd1-0ec7-4756-bafc-3ef9cbbad596","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"The highly irregular firing of cortical cells is inconsistent with temporal integration of random epsps.Journal of neuroscience, 13(1):334–350, 1993","venue":null,"work_id":null,"year":1993},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:093011d65733e0b637574e2888a4ec0aa16d3618df34777b7ca438bcd1ff6e33","observation_id":"0724d9bf-eaa2-4bfd-adbd-a51af32e3e5a","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Highly nonrandom features of synaptic connectivity in local cortical circuits.PLoS biology, 3(3):e68, 2005","venue":null,"work_id":null,"year":2005},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:679bd01b7c34d68f5603ff625c8a11d3f592ee8b5433e2e6d663430933bd567d","observation_id":"846cf88a-d6c0-4d56-a913-8e0054862d07","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Finite-time analysis of adaptive temporal difference learning with deep neural networks.Advances in Neural Information Processing Systems, 35:19592–19604, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:3a8034baaa555ebda2ce0de78909ad8a9d351e92974cfd5cf011f4828553d57c","observation_id":"139a603c-b0d7-4871-a156-6907f983b5f1","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Learning to predict by the methods of temporal differences.Machine learning, 3(1):9–44, 1988","venue":null,"work_id":null,"year":1988},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:b892e8dd6b04205ed00187aa4722232ac0f8e995a0e022ae313552702cc25e26","observation_id":"575304b0-58fb-4ef5-a30d-56f7c1907b7a","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Policy gradient methods for reinforcement learning with function approximation.Advances in neural information processing systems, 12, 1999","venue":null,"work_id":null,"year":1999},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:8ec0e79fae3ccd63e7b0b7de15222ca4958d8d11c8e83bf5c2d1024d1e71c12e","observation_id":"cc734b3d-f2a3-46f4-ba53-86ec0ba387e7","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Shared and distinct transcriptomic cell types across neocortical areas.Nature, 563(7729):72–78, 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:bf80ce335b5dbde03fca74aaf7f0099cb1adf29b7a9a135121ceaee3516d0465","observation_id":"6ecd5956-17b1-483d-a05a-d4a3bbd0008f","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"CRC press, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:b53b870af5987191014e1ed96e81ff5f7440e9026e7d51538b0dc96022191afd","observation_id":"b380d6c8-d7ab-462c-bc39-67df179d3d1d","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Applications of advances in nonlinear sensitivity analysis","venue":null,"work_id":null,"year":1981},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:a825e26f44a80d824b8739f6fc737cb06155b9f67f657e777168a0ef7fcba84b","observation_id":"20770f34-6605-4b5a-af52-17a0077fcb5f","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":"Simple statistical gradient-following algorithms for connectionist reinforcement learning.Machine learning, 8(3):229–256, 1992","venue":null,"work_id":null,"year":1992},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:b5733e1cae03f623ce66eccb3a385f608cf73a603d3fc68cf0ef60c770c93a95","observation_id":"f8bc2b22-f279-445e-9c10-d5d3b1d330bc","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T10:18:54.610927Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-06-29T10:18:54.610927Z"},"links":{"citing_paper":"/paper/2605.27946"},"observation_digest":"sha256:5003e67ffb05437f3fad2b302577b771c4d5a071ba25b659120f167bc823a26b","observation_id":"f0d67276-82ef-4543-9fcc-336ff94c2a83","resolution":{"observed_at":"2026-06-29T10:18:54.610927Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2605.27946","last_updated":"2026-05-27T04:37:04Z","latest_version":1,"primary_category":"stat.ML","snapshot_observed_at":"2026-08-17T01:34:44.990338Z","submitted_at":"2026-05-27T04:37:04Z","title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency"},"reference_resolution":{"displayed":62,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":60,"verified_exact":2,"verified_fuzzy":0},"total_outbound_references":62},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"thesis":"As of 23 August 2026, this Paper Citation Record lists 62 of 62 outbound references and 0 inbound Pith citation observations for arXiv:2605.27946."}