arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2206.05922cs.AI
收藏

从感知到程序:正则化、过参数化与摊销

From Perception to Programs: Regularize, Overparameterize, and Amortize

Cornell University(康奈尔大学)

Hao Tang, Kevin Ellis

AI总结:

本文研究神经符号程序合成,提出多任务学习、摊销推理、过参数化和程序长度可微惩罚等技术,实现感知与程序模块的端到端联合学习。

摘要:

为了将归纳推理与感知能力相结合,我们开发了用于神经符号程序合成的技术:感知输入首先由神经网络解析为低维可解释表示,随后由合成程序进行处理。我们探索了若干用于松弛该问题并通过梯度下降端到端联合学习所有模块的技术:多任务学习、摊销推理、过参数化,以及一种用于惩罚冗长程序的可微策略。总体而言,该工具箱提升了梯度引导程序搜索的稳定性,并提出了学习如何将输入感知为离散抽象,以及如何将这些抽象作为程序进行符号化处理的方法。

英文摘要:

Toward combining inductive reasoning with perception abilities, we develop techniques for neurosymbolic program synthesis where perceptual input is first parsed by neural nets into a low-dimensional interpretable representation, which is then processed by a synthesized program. We explore several techniques for relaxing the problem and jointly learning all modules end-to-end with gradient descent: multitask learning; amortized inference; overparameterization; and a differentiable strategy for penalizing lengthy programs. Collectedly this toolbox improves the stability of gradient-guided program search, and suggests ways of learning both how to perceive input as discrete abstractions, and how to symbolically process those abstractions as programs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑