Vikash Mansinghka

Vikash Mansinghka is a principal research scientist at MIT, where he leads the Probabilistic Computing Project. His group is building a new generation of probabilistic computing systems that integrate probability and randomness into the basic building blocks of software and hardware. They have discovered that this approach leads to surprising new AI capabilities, and are exploring them via a combination of academic research and entrepreneurship. They also carry out basic research on the mathematical foundations of probabilistic computation. They make our work as freely available as possible via open-source software, workshops, and online educational materials. Additionally, they collaborate with industry and non-profit partners on applications in the public interest. Mansinghka earned an SB in mathematics and computer science, MEng in computer science, and a PhD in computation, from MIT. Mansinghka co-founded two startups: Prior Knowledge and Empirical Systems, later acquired by Salesforce and Tableau, respectively.

Publications

• Saad, F. A., Freer, C. E., Rinard, M. C., Mansinghka, V. K. (2020). Optimal approximate sampling from discrete probability distributions. Proceedings of the ACM on Programming Languages, 4(POPL): 36:1–36:31.
• Lew, A. K., Cusumano-Towner, M. F., Sherman, B., Carbin, M., Mansinghka, V. K. (2020). Trace types and denotational semantics for sound programmable inference in probabilistic languages. Proceedings of the ACM on Programming Languages, 4(POPL): 19:1–19:32.
• Saad, F. A., Cusumano-Towner, M. F., Schaechtle, U., Rinard, M. C., Mansinghka, V. K. (2019). Bayesian synthesis of probabilistic programs for automatic data modeling. Proceedings of the ACM on Programming Languages, 3(POPL): 37:1–37:32.
• Felip, J., Ahuja, N., Gómez-Gutiérrez, D., Tickoo, O., Mansinghka, V. K. (2019). Real-time approximate inference for scene understanding with generative models. In Workshop on Perceptions as Generative Reasoning (co-located with NeurIPS).
• Cusumano-Towner, M. F., Saad, F. A., Lew, A. K., Mansinghka, V. K. (2019). Gen: a general-purpose probabilistic programming system with programmable inference. In Proceedings of the 40th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI). Association for Computing Machinery, 221–236.
• Kulkarni, T. D., Kohli, P., Tenenbaum, J. B., Mansinghka, V. K. (2015). Picture: A probabilistic programming language for scene perception. 28th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4390-4399.

Media

• Jun. 27, 2019: VentureBeat, MIT’s Gen programming system flattens the learning curve for AI projects.
• Jun. 26, 2019: MIT News, New AI programming language goes beyond deep learning.
• Jan. 15, 2019: MIT News, Democratizing data science.