Pith Number

pith:GZOKMZK7

pith:2026:GZOKMZK7XHVKLZN547JBDGFBGJ

not attested not anchored not stored refs resolved

Contextual Bandits for Resource-Constrained Devices using Probabilistic Learning

Amy Loutfi, Antonello Rosato, Denis Kleyko, Kevin Johansson, Marco Angioli

A probabilistic update rule lets hyperdimensional contextual bandits run on low-precision hardware while outperforming binarized versions and approaching full performance with 3 bits per component.

arxiv:2605.13346 v1 · 2026-05-13 · cs.LG

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{GZOKMZK7XHVKLZN547JBDGFBGJ}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Off-policy evaluation on standardized synthetic CB benchmarks using the Open Bandit Pipeline shows that probabilistic HD-CB consistently outperforms binarized HD-CB at equal precision, while approaching the performance of HD-CB with as few as 3 bits per component.

C2weakest assumption

That the probabilistic update rule with random subset selection and time-decaying probability preserves enough learning information to match or exceed binarized HD-CB without introducing bias or instability that would appear only in full real-world deployments beyond the synthetic benchmarks.

C3one line summary

Probabilistic HD-CB outperforms binarized HD-CB and approaches full HD-CB performance on synthetic benchmarks using as few as 3 bits per component via random partial updates with time-decaying probability.

References

30 extracted · 30 resolved · 0 Pith anchors

[1] A contextual-bandit approach to personalized news article recommendation, 2010

[2] Artwork personalization at Netflix, 2018

[3] Multiworld testing decision service: A system for experimentation, learning, and decision-making, 2016

[4] How The New York Times is experimenting with recommendation algorithms, 2019

[5] How we boosted app revenue by 10% with real-time personalization, 2020

Formal links

1 machine-checked theorem link

Receipt and verification

First computed	2026-05-18T02:44:48.336589Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

365ca6655fb9eaa5e5bde7d21198a132494a5bd764678fb6f82bb68281503516

Aliases

arxiv: 2605.13346 · arxiv_version: 2605.13346v1 · doi: 10.48550/arxiv.2605.13346 · pith_short_12: GZOKMZK7XHVK · pith_short_16: GZOKMZK7XHVKLZN5 · pith_short_8: GZOKMZK7

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/GZOKMZK7XHVKLZN547JBDGFBGJ \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 365ca6655fb9eaa5e5bde7d21198a132494a5bd764678fb6f82bb68281503516

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "c4ea35c211780cb9a5be9f995f0899ba5319534cda4de0d01ceb06c0889f1c6e",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-13T11:04:47Z",
    "title_canon_sha256": "17f21726524da37745cb916201863ae106e1d9e6a3b33cffafc63b9cbd10d493"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13346",
    "kind": "arxiv",
    "version": 1
  }
}