Sonal Gupta explores practical systems for building a conversational AI system for task-oriented queries and details a way to do more advanced compositional understanding, which can understand cross-domain queries, using hierarchical representations. Along the way, Sonal discusses a few parsers, such as transition-based parsers and sequence-to-sequence models, that can be used for compositional understanding (implemented in PyTorch).
Sonal Gupta is a research scientist at Facebook working on conversational AI systems. Previously, she developed deep learning natural language understanding models for conversational AI systems at Viv, a startup later acquired by Samsung. She holds a PhD on weakly supervised and interpretable information extraction from Stanford University and a master’s degree on combining language and vision modes for information extraction at the University of Texas at Austin.
©2019, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org