paper-with-me

Papers

Learning Program Synthesis for Integer Sequences from Scratch

2022-02-24 · Thibault Gauthier, Josef Urban

We present a self-learning approach for synthesizing programs from integer sequences. Our method relies on a tree search guided by a learned policy. Our system is tested on the On-Line Encyclopedia of Integer Sequences. There, it discovers, on its own, solutions for 27987 sequences starting from basic operators and without human-written training examples.

📄 PDF Abstract BibTeX arXiv:2202.11908

Code (3)

anonuser4213/oeis-synthesis 공식 구현
barakeel/oeis-synthesis 공식 구현 tf
anon52mi4/oeis-alien

Tasks

Program SynthesisSelf-Learning

Methods 이 논문이 사용한 방법론

Self-Learning 설명 없음

Similar Papers 제목 키워드 기반

AbstractBeam: Enhancing Bottom-Up Program Synthesis using Library Learning

2024-05-27 · Janis Zenkner, Lukas Dierkes, Tobias Sesterhenn, Chrisitan Bartelt

LambdaBeam is a state-of-the-art, execution-guided algorithm for program synthesis that utilizes higher-order functions, lambda functions, and iterative loops within a Domain-Specific Language (DSL). LambdaBeam generates…

Program Synthesis

Stepping Stones to Inductive Synthesis of Low-Level Looping Programs

2018-11-26 · Christopher D. Rosin

Inductive program synthesis, from input/output examples, can provide an opportunity to automatically create programs from scratch without presupposing the algorithmic form of the solution. For induction of general progra…

Program Synthesis

AlgoPilot: Fully Autonomous Program Synthesis Without Human-Written Programs

2025-01-11 · Xiaoxin Yin

Program synthesis has traditionally relied on human-provided specifications, examples, or prior knowledge to generate functional algorithms. Existing methods either emulate human-written algorithms or solve specific task…

Language ModelingLanguage ModellingProgram SynthesisReinforcement Learning (RL)+1

Show Your Work: Scratchpads for Intermediate Computation with Language Models

2021-11-30 · Maxwell Nye, Anders Johan Andreassen, Guy Gur-Ari, Henryk Michalewski 외

Large pre-trained language models perform remarkably well on tasks that can be done "in one pass", such as generating realistic text or synthesizing computer programs. However, they struggle with tasks that require unbou…

Opening the AI black box: program synthesis via mechanistic interpretability

2024-02-07 · Eric J. Michaud, Isaac Liao, Vedang Lad, Ziming Liu 외

We present MIPS, a novel method for program synthesis based on automated mechanistic interpretability of neural networks trained to perform the desired task, auto-distilling the learned algorithm into Python code. We tes…

Program SynthesisSymbolic Regression