• Mathematics, Computer Science
  • Published in NeurIPS 2018

Memory Augmented Policy Optimization for Program Synthesis and Semantic Parsing

@inproceedings{Liang2018MemoryAP,
  title={Memory Augmented Policy Optimization for Program Synthesis and Semantic Parsing},
  author={Chen Liang and Mohammad Norouzi and Jonathan Berant and Quoc V. Le and Ni Lao},
  booktitle={NeurIPS},
  year={2018}
}
This paper presents MAPO: a novel policy optimization formulation that incorporates a memory buffer of promising trajectories to reduce the variance of policy gradient estimates for deterministic environments with discrete actions. The formulation expresses the expected return objective as a weighted sum of two terms: an expectation over a memory of trajectories with high rewards, and a separate expectation over the trajectories outside the memory. We propose 3 techniques to make an efficient… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 22 CITATIONS

Learning to Generalize from Sparse and Underspecified Rewards

VIEW 9 EXCERPTS
CITES METHODS & BACKGROUND

Estimating Gradients for Discrete Random Variables by Sampling without Replacement

VIEW 11 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Automated Curriculum Learning for Turn-level Spoken Language Understanding with Weak Supervision

VIEW 10 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Introduction to Neural Network based Approaches for Question Answering over Knowledge Graphs

VIEW 5 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Learning Semantic Parsers from Denotations with Latent Structured Alignments and Abstract Programs

VIEW 9 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

A Discrete Hard EM Approach for Weakly Supervised Question Answering

VIEW 4 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

A Sketch-Based System for Semantic Parsing

VIEW 1 EXCERPT
CITES BACKGROUND

References

Publications referenced by this paper.
SHOWING 1-10 OF 63 REFERENCES

Compositional Semantic Parsing on Semi-Structured Tables

VIEW 17 EXCERPTS
HIGHLY INFLUENTIAL