Pith. sign in

REVIEW 5 cited by

Continual Backprop: Stochastic Gradient Descent with Persistent Randomness

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2108.06325 v3 pith:V4J6HRH7 submitted 2021-08-13 cs.LG

Shibhansh Dohare , Richard S. Sutton , A. Rupam Mahmood This is my paper · ORCID
classification cs.LG
keywords backpropcontinuallearningdescentgradientalgorithmcontinuallystochastic
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The Backprop algorithm for learning in neural networks utilizes two mechanisms: first, stochastic gradient descent and second, initialization with small random weights, where the latter is essential to the effectiveness of the former. We show that in continual learning setups, Backprop performs well initially, but over time its performance degrades. Stochastic gradient descent alone is insufficient to learn continually; the initial randomness enables only initial learning but not continual learning. To the best of our knowledge, ours is the first result showing this degradation in Backprop's ability to learn. To address this degradation in Backprop's plasticity, we propose an algorithm that continually injects random features alongside gradient descent using a new generate-and-test process. We call this the \textit{Continual Backprop} algorithm. We show that, unlike Backprop, Continual Backprop is able to continually adapt in both supervised and reinforcement learning (RL) problems. Continual Backprop has the same computational complexity as Backprop and can be seen as a natural extension of Backprop for continual learning.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. To Retain or to Adapt? Generalizing Continual Learning

    stat.ML 2026-07 accept novelty 7.5 of 10

    When retention induces positive stationary bias, historical knowledge becomes an optimization liability beyond a closed-form Critical Task Duration; Predictive CL with finite windows can beat both full Joint-Task and ...

  2. Beyond Single-Model Optimization: Preserving Plasticity in Continual Reinforcement Learning

    cs.LG 2026-04 unverdicted novelty 7.0 of 10

    TeLAPA preserves behaviorally diverse policy neighborhoods in a shared latent space, improving MiniGrid continual RL transfer, revisit recovery, and retention over single-model preservation.

  3. Calibrated Partial Resets: Preventing Policy Collapse in Continual Reinforcement Learning

    cs.LG 2026-07 conditional novelty 5.5 of 10

    Utility-scaled partial neuron resets prevent policy collapse in long-horizon continual RL while matching or beating binary-reset and uniform-decay baselines on several benchmarks.

  4. Reinitializing weights vs units for maintaining plasticity in neural networks

    cs.NE 2025-07 conditional novelty 5.0 of 10

    Selective weight reinitialization, which resets the least useful weights, maintains plasticity in small and layer-normalized networks where unit-level reinitialization methods fail.

  5. Optimizers Qualitatively Alter Solutions And We Should Leverage This

    cs.LG 2025-07 conditional novelty 4.0 of 10

    Deep learning optimizers should be designed to induce desired solution properties, not just convergence speed; different optimizers demonstrably land in qualitatively different minima.

Pith tools