Building Intelligent Probabilistic Systems

In the HIPS group, we are interested in building intelligent algorithms. What makes a system intelligent? Our philosophy is that "intelligence" means making decisions under uncertainty, adapting to experience, and discovering structure in high-dimensional noisy data. The unifying theme for research in these areas is developing new approaches to statistical inference: uncovering the coherent structure that we cannot directly observe and using it for exploration and to make decisions or predictions. We develop new models for data, new tools for performing inference, and new computational structures for representing knowledge and uncertainty.

Recent News

MICMAT: Python Scientific Computing on Intel Xeon Phi

 

Oren Rippel has recently released the first version of his new Python library MICMAT. This library has a similar functionality to cudamat, but is focused on Intel's newXeon Phi (MIC) architecture, rather than nVidia's CUDA.

"Firefly Monte Carlo" Wins Best Paper at UAI

 

Dougal Maclaurin's paper Firefly Monte Carlo: Exact MCMC with Subset of Data has won the Microsoft Best Paper Award at this year's Conference on Uncertainty in Artificial Intelligence (UAI). Congrats Dougal!

Five Papers at ICML 2014

 

The HIPS group co-authored five papers to appear at this year's International Conference on Machine Learning (ICML).

Netflix Using Spearmint for Bayesian Optimization

 

As reported by Wired magazine and on the Netflix tech blog. Netflix has been experimenting with deep learning tools for making recommendations. Moreover, they've been using our software Spearmint to set the hyperparameters with a cluster of machines on Amazon EC2.

Recent Publications