arXiv subjects

Jillian Tang

Publications and source records attributed to Jillian Tang.

3 recordsLinked to original sources

arXiv2022

Neural Generation Meets Real People: Building a Social, Informative Open-Domain Dialogue Agent

We present Chirpy Cardinal, an open-domain social chatbot. Aiming to be both informative and conversational, our bot chats with users in an authentic, emotionally intelligent way. By integrating controlled neural generation with scaffolded, hand-written dialogue, we let both the user and bot take turns driving the conversation, producing an engaging and socially fluent experience. Deployed in the fourth iteration of the Alexa Prize Socialbot Grand Challenge, Chirpy Cardinal handled thousands of conversations per day, placing second out of nine bots with an average user rating of 3.58/5.

cs.CL↗

arXiv2019

SGVAE: Sequential Graph Variational Autoencoder

Generative models of graphs are well-known, but many existing models are limited in scalability and expressivity. We present a novel sequential graphical variational autoencoder operating directly on graphical representations of data. In our model, the encoding and decoding of a graph as is framed as a sequential deconstruction and construction process, respectively, enabling the the learning of a latent space. Experiments on a cycle dataset show promise, but highlight the need for a relaxation of the distribution over node permutations.

cs.LG↗

arXiv2019

Some t-tests for N-of-1 trials with serial correlation

N-of-1 trials allow inference between two treatments given to a single individual. Most often, clinical investigators analyze an individual's N-of-1 trial data with usual t-tests or simple nonparametric methods. These simple methods do not account for serial correlation in repeated observations coming from the individual. Existing methods accounting for serial correlation require simulation, multiple N-of-1 trials, or both. Here, we develop t-tests that account for serial correlation in a single individual. The development includes effect size and precision calculations, both of which are useful for study planning. We then evaluate and compare their Type I and II errors and interval estimators to those of usual t-tests analogues via Monte Carlo simulation. The serial t-tests clearly outperform the usual t-tests commonly used in reporting N-of-1 results. Examples from N-of-1 clinical trials in fibromyalgia patients and from a behavioral health setting exhibit how accounting for serial correlation can change inferences. These t-tests are easily implemented and more appropriate than simple methods commonly used; however, caution is needed when analyzing only a few observations. Keywords: Autocorrelation; Cross-over studies; Repeated measures analysis; Single-case experimental design; Time-series

stat.ME↗