Seminar

Aarti Malhotra, PhD candidate
David R. Cheriton School of Computer Science

We investigate a deep learning approach to classify emotions in group-level photos into three categories: (1) positive; (2) neutral; and (3) negative. 

Joshua Jung, PhD candidate
David R. Cheriton School of Computer Science

Recent work established that it is possible for human artists to encode information into hand-drawn markers, but it is difficult to do when simultaneously maintaining aesthetic quality. We present two methods for relieving the mental burden associated with encoding, while allowing an artist to draw as freely as possible. 

Shrinu Kushagra, PhD candidate
David R. Cheriton School of Computer Science

Data de-duplication is the task of detecting multiple records that correspond to the same real-world entity in a database. In this work, we view de-duplication as a clustering problem. We introduce a framework which we call promise correlation clustering. Given a complete graph \(G\) with the edges labeled \(0\) and \(1\), the goal is to find a clustering that minimizes the number of \(0\) edges within a cluster plus the number of \(1\) edges across different clusters (or correlation loss). The optimal clustering can also be viewed as a complete graph \(G^*\) with edges corresponding to points in the same cluster being labeled \(0\) and other edges being labeled \(1\).

Aishwarya Agrawal, PhD candidate
School of Interactive Computing, Georgia Tech

In this talk, I will present our work on a multi-modal AI task called Visual Question Answering (VQA) — given an image and a natural language question about the image (e.g., “What kind of store is this?”, “Is it safe to cross the street?”), the machine’s task is to automatically produce an accurate natural language answer (“bakery”, “yes”).

Nabiha Asghar, PhD candidate
David R. Cheriton School of Computer Science

Most of the existing neural conversational models process natural language primarily on a lexico-syntactic level, thereby ignoring one of the most crucial components of human-to-human dialogue: its affective content. We take a step in this direction by proposing three novel ways to incorporate affective/emotional aspects into long short term memory (LSTM) encoder-decoder neural conversation models: (1) affective word embeddings, which are cognitively engineered, (2) affect-based objective functions that augment the standard cross-entropy loss, and (3) affectively diverse beam search for decoding.

Andrew Delong, Head of Computational Research
Deep Genomics

Genomics focuses on the sequences in our genomes and how they encode for function in our cells. Predicting how sequences will be interpreted by the cell is important for identifying disease-causing mutations and for designing therapies.