Pith Number

pith:QT4KCTXH

pith:2026:QT4KCTXHANQ2O4DGZHDYSAXCG7

not attested not anchored not stored refs resolved

Domain-Independent Game Abstraction using Word Embedding Techniques

Juho Kim, Tuomas Sandholm

Word embeddings trained on gameplay data can cluster actions to create effective abstractions for large games without domain-specific knowledge.

arxiv:2605.15543 v1 · 2026-05-15 · cs.GT · cs.AI

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{QT4KCTXHANQ2O4DGZHDYSAXCG7}

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

Experimental results demonstrate that our proposed game abstraction technique is effective, although it does not outperform specialized algorithms tailored to specific games.

C2weakest assumption

That vector similarities learned from raw action sequences in gameplay data will group actions in a way that preserves strategic value for the purpose of abstraction, without additional game-specific features or labels.

C3one line summary

Uses word embeddings on action sequences from gameplay data to produce domain-independent game abstractions via vector clustering, with experiments showing effectiveness but no outperformance over specialized methods.

References

36 extracted · 36 resolved · 0 Pith anchors

[1] D. Billings, N. Burch, A. Davidson, R. Holte, J. Schaeffer, T. Schauenberg, and D. Szafron. Approximating game-theoretic optimal strategies for full-scale poker. InProceedings of the International Joi 2003

[2] N. Brown and T. Sandholm. Superhuman AI for heads-up no-limit poker: Libratus beats top professionals.Science, 359(6374):418–424, 2018 2018

[3] N. Brown and T. Sandholm. Superhuman AI for multiplayer poker.Science, 365(6456):885–890, 2019 2019

[4] N. Brown, S. Ganzfried, and T. Sandholm. Hierarchical abstraction, distributed equilibrium computation, and post-processing, with application to a champion no-limit Texas hold’em agent. InProceedings 2015

[5] R. Carlson, J. Bauer, and C. D. Manning. A new pair of GloVes, 2025 2025

Formal links

1 machine-checked theorem link

Receipt and verification

First computed	2026-05-20T00:01:04.434171Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

84f8a14ee70361a77066c9c78902e237c885a76e6df8b26900d4a12d45f197f6

Aliases

arxiv: 2605.15543 · arxiv_version: 2605.15543v1 · doi: 10.48550/arxiv.2605.15543 · pith_short_12: QT4KCTXHANQ2 · pith_short_16: QT4KCTXHANQ2O4DG · pith_short_8: QT4KCTXH

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/QT4KCTXHANQ2O4DGZHDYSAXCG7 \
  | 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: 84f8a14ee70361a77066c9c78902e237c885a76e6df8b26900d4a12d45f197f6

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "67a998ff300182a8d85cd021d183663951c4a29c08f39164b4c918ac02173caa",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.GT",
    "submitted_at": "2026-05-15T02:30:36Z",
    "title_canon_sha256": "6292d94238f2ff7daefe18a9097c2bc48872d2393be4364f67478eed6665778f"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.15543",
    "kind": "arxiv",
    "version": 1
  }
}