Timezone: »

Memory Augmented Policy Optimization for Program Synthesis and Semantic Parsing
Chen Liang · Mohammad Norouzi · Jonathan Berant · Quoc V Le · Ni Lao

Wed Dec 05 12:40 PM -- 12:45 PM (PST) @ Room 220 CD

We present Memory Augmented Policy Optimization (MAPO), a simple and novel way to leverage a memory buffer of promising trajectories to reduce the variance of policy gradient estimate. MAPO is applicable to deterministic environments with discrete actions, such as structured prediction and combinatorial optimization tasks. We express the expected return objective as a weighted sum of two terms: an expectation over the high-reward trajectories inside the memory buffer, and a separate expectation over trajectories outside the buffer. To make an efficient algorithm of MAPO, we propose: (1) memory weight clipping to accelerate and stabilize training; (2) systematic exploration to discover high-reward trajectories; (3) distributed sampling from inside and outside of the memory buffer to scale up training. MAPO improves the sample efficiency and robustness of policy gradient, especially on tasks with sparse rewards. We evaluate MAPO on weakly supervised program synthesis from natural language (semantic parsing). On the WikiTableQuestions benchmark, we improve the state-of-the-art by 2.6%, achieving an accuracy of 46.3%. On the WikiSQL benchmark, MAPO achieves an accuracy of 74.9% with only weak supervision, outperforming several strong baselines with full supervision. Our source code is available at https://goo.gl/TXBp4e

Author Information

Chen Liang (Google Brain)
Mohammad Norouzi (Google Brain)
Jonathan Berant (Tel Aviv University)
Quoc V Le (Google)
Ni Lao (Mosaix.ai)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors