Christopher Tosh

Memorial Sloan Kettering Cancer Center
Department of Epidemiology and Biostatistics

Academic CV
Google Scholar
E-mail: christopher DOT j DOT tosh AT gmail DOT com

Short biography

I am a postdoc in the Department of Epidemiology and Biostatistics at Memorial Sloan Kettering Cancer Center. My current interests are on problems arising in interactive learning, representation learning, and robust learning, particularly with applications to cancer research. I received my PhD from UC San Diego, where I was advised by Sanjoy Dasgupta.

Selected papers

M. Simchowitz, C. Tosh, A. Krishnamurthy, D. Hsu, T. Lykouris, M. Dudík, and R. E. Schapire. Bayesian decision-making under misspecified priors with applications to meta-learning.
Neural Information Processing Systems (NeurIPS), 2021. [Full version]

C. Tosh, A. Krishnamurthy and D. Hsu. Contrastive learning, multi-view redundancy, and linear models.
Conference on Algorithmic Learning Theory (ALT), 2021. [Full version]

C. Tosh, A. Krishnamurthy and D. Hsu. Contrastive estimation reveals topic posterior information to linear models.
Journal of Machine Learning Research, 22(281):1−31, 2021.

S. Dasgupta and C. Tosh. Expressivity of expand-and-sparsify representations.
Preprint, 2020.

C. Tosh and D. Hsu. Diameter-based interactive structure discovery.
Twenty-third International Conference on Artificial Intelligence and Statistics (AISTATS), 2020. [Full version]

W. Tansey, C. Tosh, and D. M. Blei. A Bayesian model of dose-response for cancer drug studies.
Preprint, 2019.

C. Tosh and S. Dasgupta. The relative complexity of maximum likelihood estimation, MAP estimation, and sampling.
Conference on Learning Theory (COLT), 2019.

C. Tosh and S. Dasgupta. Interactive structure learning with structural query-by-committee.
Neural Information Processing Systems (NeurIPS), 2018. [Full version]

C. Tosh. Algorithms for statistical and interactive learning tasks.
PhD Thesis, 2018.

C. Tosh and S. Dasgupta. Maximum likelihood estimation for mixtures of spherical Gaussians is NP-hard.
Journal of Machine Learning Research, 18(175):1-11, 2018.

C. Tosh and S. Dasgupta. Diameter-based active learning.
Thirty-fourth International Conference on Machine Learning (ICML), 2017. [Full version]

C. Tosh. Mixing Rates for the alternating Gibbs sampler over Restricted Boltzmann Machines and friends.
Thirty-third International Conference on Machine Learning (ICML), 2016. [Full version]

C. Tosh and S. Dasgupta. Lower bounds for the Gibbs sampler over mixtures of Gaussians.
Thirty-first International Conference on Machine Learning (ICML), 2014. [Full version]