Sounding Board is a conversational agent that engages users through thoughtful and informative discussion.
Our team represents University of Washington natural language processing (NLP) with PhD students Hao F. and Hao C. from Electrical Engineering (EE) and Elizabeth, Ariel, and Maarten from Computer Science & Engineering (CSE). We're working with EE Professor Mari Ostendorf and CSE Professors Yejin Choi and Noah Smith. The students have expertise in language generation, deep learning, reinforcement learning, psychologically-infused NLP, and human-AI collaboration. Together, they aim to create a conversational agent that transforms the way people interact with everyday home devices.
Hao F. - Team leader
I'm interested in developing and implementing state-of-the-art methods which can impact to real-world applications. I've been working on neural network models for spoken and language processing since 2013, and did several exciting projects including automatic image captioning, natural language understanding, community endorsement analysis in online discussion forums, and low-resource language modeling. My thesis topic is about automatic rebuttal generation in debates. I have also developed our own C++ neural network library with my friend, which becomes a handy tool for our scientific research.
Ari is a 1st year graduate student studying Natural Language Generation under Yejin Choi. His current work explores methods of conditioning text generation on different types of context, including conditioning on a fully-formed sentence in order to refine it. His interests include combining Reinforcement Learning and Deep Learning, goal-oriented text generation, and situated language understanding.
I'm a second year PhD student working with Noah Smith and am part of the Natural Language Processing group. I'm interested in human-computer collaboration and language generation, particularly in generating language that promotes creativity in users.
I'm a 3rd year PhD student working with Mari Ostendorf. My research interests center around machine learning and natural language processing. For machine learning, I have been working on novel deep learning models, tensor-based learning and optimization methods. For NLP, I'm interested in parsing, social media, text generation and healthcare analytics. Most of my research work results in open-source software and top-tier publications.
I'm a second year PhD student in Computer Science, working with Noah Smith and Yejin Choi. I'm particularly interested in the intersection of NLP and Psychology/Social Science. I'm a strong believer in NLP for Social Good!
Mari Ostendorf - Faculty advisor
Mari received her PhD in Electrical Engineering from Stanford University. Currently, she is a Professor of Electrical Engineering at the University of Washington, with adjunct positions in Linguistics and Computer Science & Engineering. She holds an Endowed Professorship in System Design Methodology. Her research interests are in speech and language processing, particularly dynamic data-driven models. She has held visiting appointments at ATR, Karlsruhe University, Edinburgh and Macquarie University. She is a fellow of the IEEE and ISCA, a former Fulbright Scholar, and IEEE Distinguished Lecturer.