ACES: Generating diverse programming puzzles with autotelic language models and semantic descriptors - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Preprints, Working Papers, ... Year : 2023

ACES: Generating diverse programming puzzles with autotelic language models and semantic descriptors

Julien Pourcel
Cédric Colas
Pierre-Yves Oudeyer

Abstract

Finding and selecting new and interesting problems to solve is at the heart of curiosity, science and innovation. We here study automated problem generation in the context of the open-ended space of python programming puzzles. Existing generative models often aim at modeling a reference distribution without any explicit diversity optimization. Other methods explicitly optimizing for diversity do so either in limited hand-coded representation spaces or in uninterpretable learned embedding spaces that may not align with human perceptions of interesting variations. With ACES (Autotelic Code Exploration via Semantic descriptors), we introduce a new autotelic generation method that leverages semantic descriptors produced by a large language model (LLM) to directly optimize for interesting diversity, as well as few-shot-based generation. Each puzzle is labeled along 10 dimensions, each capturing a programming skill required to solve it. ACES generates and pursues novel and feasible goals to explore that abstract semantic space, slowly discovering a diversity of solvable programming puzzles in any given run. Across a set of experiments, we show that ACES discovers a richer diversity of puzzles than existing diversity-maximizing algorithms as measured across a range of diversity metrics. We further study whether and in which conditions this diversity can translate into the successful training of puzzle solving models.
Fichier principal
Vignette du fichier
2310.10692.pdf (3.37 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04372580 , version 1 (04-01-2024)

Licence

Attribution

Identifiers

  • HAL Id : hal-04372580 , version 1

Cite

Julien Pourcel, Cédric Colas, Pierre-Yves Oudeyer, Laetitia Teodorescu. ACES: Generating diverse programming puzzles with autotelic language models and semantic descriptors. 2024. ⟨hal-04372580⟩

Collections

INRIA INRIA2
25 View
9 Download

Share

Gmail Facebook X LinkedIn More