Generative probabilistic programming: applications and new ideas

Channel:
Microsoft Research
Subscribers:
348,000
Published on ● Video Link: https://www.youtube.com/watch?v=M_vG_6pq0XM


Duration: 57:06
969 views
13

Probabilistic programming has recently attracted much attention in Computer Science and Machine Learning communities. I will briefly demonstrate two generative probabilistic graphics programs (models), which I contributed to develop. Then I will present ideas on two research directions I am interested in pursuing: a path to scaling up general-purpose approximate inference in probabilistic programs using parallelism, and a path to automatic programming via general-purpose approximate inference. This is based on joint work with Frank Wood, Vikash Mansinghka, Tejas Kulkarni, Daniel Selsam, Joshua Tenenbaum, et al.



Other Videos By Microsoft Research


2016-06-22Posner Lecture: Probabilistic Machine Learning - Foundations and Frontiers
2016-06-22Machine Learning Course - Lecture 6
2016-06-22Interactive and Interpretable Machine Learning Models for Human Machine Collaboration
2016-06-22Oral Session: Sampling from Probabilistic Submodular Models
2016-06-22Invited Talk: Learning with Intelligent Teacher: Similarity Control and Knowledge Transfer
2016-06-22Torch Future Plans
2016-06-22Machine Learning Course - Lecture 3
2016-06-22Greedy Transition-Based Dependency Parsing with Stack-LSTMs
2016-06-22Invited Talk: Incremental Methods for Additive Cost Convex Optimization
2016-06-22How Much Information Does a Human Translator Add to the Original and Multi-Source Neural Translation
2016-06-22Generative probabilistic programming: applications and new ideas
2016-06-22Making Small Spaces Feel Large: Practical Illusions in Virtual Reality
2016-06-22Python+Machine Learning tutorial - Working with textual data
2016-06-22Coping with the Intractability of Graphical Models
2016-06-22Symposium: Algorithms Among Us - Intro & Tom Dietterich
2016-06-22Gates Foundation Presents: Crucial Areas of Fintech Innovation for the Bottom of the Pyramid
2016-06-22Symposium: Brains, Minds and Machines - Andrew Saxe
2016-06-22Machine Learning in Health Care
2016-06-22Human factors of software updates
2016-06-22Locality-Sensitive Hashing and Beyond
2016-06-22Oral Session: Competitive Distribution Estimation: Why is Good-Turing Good


Tags:
microsoft research
machine learning
deep neural networks