Learning without Annotation
Supervised learning has worked very well in practice but yet they require exhaustive labelled data, posing a barrier to widespread adoption. My research works towards a holy grail of unsupervised learning that can either learn without annotation completely or with noisy and easy to obtain labels. I like to think about "representation learning" algorithms that learn to capture the inherent structure of the data itself (e.g. the distribution or invariances in properties), or multimodality, where an assortment of images, text, videos, audio clips, etc each provides a stronger learning signal we could use to build a richer and more abstract representation.
Preprints
ArXiv
Submitted to JMLR
Submitted to JAIR
Optimizing for Interpretability in Deep Neural Networks with Simulable Decision Trees
Mike Wu, Sonali Parbhoo, Michael C. Hughes, Volker Roth, Finale Doshi-Velez
PDFPublications
ICLR 2021
Viewmaker Networks: Learning Views for Unsupervised Representation Learning
Alex Tamkin, Mike Wu, Noah Goodman
PDFICLR 2021
Conditional Negative Sampling for Contrastive Learning of Visual Representations
Mike Wu, Milan Mosse, Chengxu Zhuang, Daniel Yamins, Noah Goodman
PDFArXiv 2020 (used for CS224S)
HarperValleyBank: A Domain-Specific Spoken Dialog Corpus
Mike Wu, Jonathan Nafziger, Anthony Scodary, Andrew Maas
PDFEDM 2020
Variational Item Response Theory: Fast, Accurate, and Expressive
Mike Wu, Richard L. Davis, Benjamin W. Domingue, Chris Piech, Noah Goodman
PDF Oral Best PaperArXiv 2019
Generative Grading: Neural Approximate Parsing for Automated Student Feedback
Ali Malik (*), Mike Wu (*), Vrinda Vasavada, Jinpeng Song, John Mitchell, Noah Goodman, Chris Piech
PDFCOBB 2019
Pragmatic inference and Visual Abstraction Enable Contextual Flexibility during Visual Communication
Judith Fan, Robert X.D. Hawkins, Mike Wu, Noah Goodman
PDFAISTATS 2019
Differentiable Antithetic Sampling for Variance Reduction in Stochastic Variational Inference
Mike Wu, Noah Goodman, Stefano Ermon
PDFAAAI 2019
Zero Shot Learning for CodeEducation: Rubric Sampling with Deep Learning Inference
Mike Wu, Milan Mosse, Noah Goodman, Chris Piech
PDF Oral Best Student PaperNeurIPS 2018
PROBPROG 2018
Spreadsheet Probabilistic Programming
Mike Wu, Yura Perov, Frank Wood, Hongseok Yang, William Smith
PDFAMIA CRI 2017
Predicting Intervention Onset in the ICU with Switching Statespace Models
Marzyeh Ghassemi, Mike Wu, Michael C. Hughes, Finale Doshi-Velez
PDF Oral Best Paper FinalistJAMIA 2016
Understanding Vassopressor Intervention and Weaning: Risk Prediction in a Public Heterogeneous Clinical Time Series Database
Mike Wu, Marzyeh Ghassemi, Mengling Feng, Leo Anthony Celi, Peter Szolovitz, Finale Doshi-Velez
PDFConference Abstracts
VSS 2018
Modeling Contextual Flexibility in Visual Communication
Judith Fan, Robert X.D. Hawkins, Mike Wu, Noah Goodman
AAS 2018, Undergrad Thesis
Investigating the Cosmic Web with Topological Data Analysis
Jessi Cisewski-Kehe, Mike Wu, Brittany Fasy, Wojciech Hellwing, Mark Lovell, Alessandro Rinaldo, Larry Wasserman
XSEDE 2011, Siemens Semifinalist 2012, ISEF Finalist 2011: 3rd place
Position and Vector Detection of Blind Spot motion with the Horn-Schunck Optical Flow
Stephen Yu, Mike Wu
XSEDE Best Student Poster