Sriram Gopalakrishnan, Ph.D.

Sriram Gopalakrishnan, Ph.D.

AI Research Lead at J.P. Morgan Chase | Sequential-Decision Making Problems, Human-AI Interaction, & Generative AI

Learning Task Hierarchies Using Statistical Semantics and Goal Reasoning

Published in AI Communications Journal, 31(2): 119-137 (2018), 2018

Recommended citation: Gopalakrishnan, S., Muñoz-Avila, H., & Kuter, U. (2018). Learning Task Hierarchies Using Statistical Semantics and Goal Reasoning. AI Communications, 31(2), 119-137. https://content.iospress.com/articles/ai-communications/aic170752

Abstract

Constructing Hierarchical Task Networks (HTNs) by hand is notoriously knowledge-intensive. We present Word2HTN, a novel framework leveraging distributional semantic models (word embeddings) over action descriptors to automatically induce method hierarchies from flat execution traces.

View Publication

Recommended citation: Gopalakrishnan, S., Muñoz-Avila, H., & Kuter, U. (2018). Learning Task Hierarchies Using Statistical Semantics and Goal Reasoning. AI Communications, 31(2), 119-137.