Karel

The 'Karel' dataset/benchmark contains environments for evaluating programmatic reinforcement learning policies, specifically focusing on the effectiveness and efficiency of program generation and optimization methods.

Papers using Karel (9)

Evaluating ChatGPT and GPT-4 for Visual Programming2023 · 18 citesSynthesize, Execute and Debug: Learning to Repair for Neural Program Synthesis2020 · 12 citesNeural Program Meta-Induction2017 · 1 citesPLANS: Robust Program Learning from Neurally Inferred Specifications2020 · 1 citesAdaptive Scaffolding in Block-Based Programming via Synthesizing New Tasks as Pop Quizzes2023 · 1 citesIReEn: Reverse-Engineering of Black-Box Functions via Iterative Neural Program Synthesis2020Latent Execution for Neural Program Synthesis2021Hierarchical Programmatic Reinforcement Learning via Learning to Compose Programs2023Synthesizing Programmatic Reinforcement Learning Policies with Large Language Model Guided Search2024