Dawen Liang

Email: Github: @dawenl I am a research scientist at Netflix. I completed my Ph.D. in the Electrical Engineering Department at Columbia University, as part of the LabROSA, working with Professor Dan Ellis and Professor David Blei. My CV is available here, as of Sep 2020. Prior to that, I completed my master's degree in Music and Technology program at Carnegie Mellon University, working with Professor Roger Dannenberg in the School of Computer Science, where I was part of the Computer Music Group. I received my bachelor's degree in Computer Science at Fudan University.

Education

• Ph.D., Electrical Engineering, Columbia University, June 2016

• M.S., Music and Technology, Carnegie Mellon University, May 2012

• B.S., Computer Science, Fudan University, Jun 2010

Research

• Probabilistic latent variables models and applications to recommender systems.

• Bandit, reinforcement learning, and causal inference.

Experience

• Research Scientist, Product Machine Learning Research, Netflix, July 2016 – present

• Recommendation Systems Scientist Intern, Playlist Team, Pandora, May 2015 – Aug 2015

  • Investigate hybrid approaches to collaborative filtering. Mentors: Erik Schmidt and Keki Burjorjee.

• Research Intern, Adobe Creative Technology Laboratory, Adobe Inc., May 2013 – Aug 2013, May 2014 – Aug 2014

  • Bayesian hierarchical model of audio. Mentors: Matt Hoffman and Gautham Mysore.

Papers

For the most up-to-date information please see my Google Scholar page.

Miscellaneous

Here are some notes regarding some topics I found interesting. I wrote these just for fun (and I think they will be useful sometime in the future). Therefore, there might be mistakes/errors in these notes. If you find any problem, please let me know.

• A proof that Maximum Likelihood Estimation (MLE) is biased for variance estimator.

• A note that shows some useful properties for matrix derivative.

• Some technical details for the Expectation Maximization (EM) algorithm.