Victoria Dean

I am a PhD student in the Robotics Institute at Carnegie Mellon University advised by Abhinav Gupta. I am interested in self-supervised learning and how it can enable robots to autonomously perform complex tasks.

I completed my B.S. in Computer Science and Engineering at MIT, where I did research with Antonio Torralba and Carl Vondrick. I have also spent time at Waymo, Deep Genomics, Counsyl, Google, and Coursera.

vdean@cs.cmu.eduCVGithub

Research

CaRL: Combining Imitation Learning with Reinforcement Learning

Combined imitation learning with reinforcement learning, which allows us to both learn from observed behavior and generalize beyond it. Applied the idea to a trajectory generation task. Waymo, spring 2018

Onboard Text Detection

Developed and deployed a new onboard text detection system that allows cars to recognize a set of important words to disambiguate a number of tricky situations. Used an accurate but expensive Google OCR system to train a low-latency onboard model. Waymo, fall 2017

Deep Learning for Branch Point Selection in RNA Splicing

Built BRANCHR, a state-of-the-art model for predicting branch point selection, a crucial step in RNA splicing. Convolutional neural network built with TensorFlow. Work accepted to WIML and MLCB at NeurIPS 2016. Selected for oral presentation at MLCB. Deep Genomics, summer 2016 [paper]

Understanding Personality with Deep Convolutional Neural Networks

Created personality dataset by having Mechanical Turk users take personality quizzes for characters in Hollywood movies. Trained models to use movie clips to classify Skip-Thought Vectors, which are correlated with character personality. Presented poster at EECScon 2016. MIT Computer Vision group, spring 2016 [poster]

Predicting the Future: Generative Models for Video

Experimented with CNN models to predict a frame given past frames. Presented poster at SuperUROP poster session. MIT Computer Vision group, Fall 2015 [poster]

Automated Search for Lyman Alpha Emitters

Developed pattern matching software to search spectra for distant galaxies. Presented poster at 2013 American Astronomical Society meeting. Intel Science Talent Search semifinalist. UC Santa Cruz, 2011-2012 [paper][poster 1][poster 2]

Teaching

FIRST Robotics Competition - Gatorbotics
Head Coach, 2017 - 2018

MIT Introduction to Deep Learning - 6.S191
Lecturer, January 2017 [video]

MIT Global Teaching Labs - Istituto Tecnico Industriale Statale Tullio Buzzi di Prato
Instructor, January 2016

MIT SWE - #HelloWorld
Instructor, Spring 2015 - Fall 2016 [slides]

MIT Introduction to Electrical Engineering and Computer Science - 6.01
Student Lab Assistant, Spring 2014

Khan Academy - Discovery Lab
Teaching Assistant, Summer 2012

Other

Outside of research, I enjoy experimental baking, swing dancing, and rowing.

I co-founded Code for Good, an MIT student group that works with Boston-area nonprofits on technical projects.

Languages: English (native), Mandarin (conversational), Spanish (beginner)