ProTo: Program-Guided Transformer for Program-Guided Tasks

Zelin Zhao, Karan Samel, Binghong Chen, Le Song

Keywords: program-guided tasks, learning to execute, program-guided agent, neural symbolic reasoning

TL;DR: We formulate program-guided tasks and present the program-guided transformer for program-guided tasks.

Abstract: Programs, consisting of semantic and structural information, play an important role in the communication between humans and agents. Towards learning general program executors to unify perception, reasoning, and decision making, we formulate program-guided tasks which require learning to execute a given program on the observed task specification. Furthermore, we propose Program-Guided Transformers (ProTo), which integrates both semantic and structural guidance of a program by leveraging cross-attention and masked self-attention to pass messages between the specification and routines in the program. ProTo executes a program in a learned latent space and enjoys stronger representation ability than previous neural-symbolic approaches. We demonstrate that ProTo significantly outperforms the previous state-of-the-art methods on GQA visual reasoning and 2D Minecraft policy learning datasets. Additionally, ProTo demonstrates better generalization to unseen, complex, and human-written programs.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/sjtuytc/Neurips21-ProTo-Program-guided-Transformers-for-Program-guided-Tasks