Search
mlsplogo MLSP2016
IEEE International Workshop on
Machine Learning for Signal Processing

September 13-16, 2016  Vietri sul Mare, Salerno, Italy

Probabilistic Programming for Augmented Intelligence
Vikash K. Mansinghka Research Scientist Vikash K. Mansinghka
Computer Science and Artificial Intelligence Laboratory & Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology, Boston, USA
Group homepage

Biography:

Vikash Mansinghka is a research scientist at MIT, where he leads the Probabilistic Computing Project. Vikash holds S.B. degrees in Mathematics and in Computer Science from MIT, as well as an M.Eng. in Computer Science and a PhD in Computation. He also held graduate fellowships from the National Science Foundation and MIT's Lincoln Laboratory. His PhD dissertation on natively probabilistic computation won the MIT George M. Sprowls dissertation award in computer science, and his research on the Picture probabilistic programming language won an award at CVPR. He co-founded a venture-backed startup based on this research that was acquired by Salesforce.com, was an advisor to Google DeepMind, and is a co-founder of Empirical Systems, a new venture-backed AI startup aimed at improving the credibility and transparency of statistical inference. He served on DARPA's Information Science and Technology advisory board from 2010-2012, and currently serves on the editorial boards for the Journal of Machine Learning Research and the journal Statistics and Computation.

Abstract:

If people could communicate with and interactively modify the behavior of AI systems, both people and machines could behave more intelligently. Unfortunately, most AI systems are black boxes designed to solve a single narrowly defined problem, such as chess or face recognition or click prediction, and adjusting their behavior requires deep technical expertise. In this talk, I will describe progress towards more transparent and flexible AI systems capable of augmenting rather than just replacing human intelligence, building on the emerging field of probabilistic programming. Probabilistic programming draws on probability theory, programming languages, and system software to provide concise, expressive languages for modeling and general-purpose inference engines that both humans and machines can use.

This talk focuses on BayesDB and Picture, domain-specific probabilistic programming platforms being developed by my research group, aimed at augmenting intelligence in the fields of data science and computer vision, respectively. BayesDB, which is open source and in use by organizations like the Bill & Melinda Gates Foundation and JPMorgan, lets users who lack statistics training understand the probable implications of data by writing queries in a simple, SQL-like language. Picture, a probabilistic language being developed in collaboration with Microsoft, lets users solve hard computer vision problems such as inferring 3D models of faces, human bodies and novel generic objects from single images by writing short (<50 line) computer graphics programs that generate and render random scenes. Unlike bottom-up vision algorithms, Picture programs build on prior knowledge about scene structure and produce complete 3D wireframes that people can manipulate using ordinary graphics software. This talk will also briefly illustrate the fundamentals of probabilistic programming using Venture, an interactive platform suitable for teaching and applications in fields ranging from statistics to robotics, and concludes with a summary of current and future research directions.


Probabilistic graphical models for bayesian learning of state flow models in non stationary environments
Carlos Regazzoni Prof. Carlo Regazzoni
Cognitive Telecommunications Systems at DITEN
University of Genova, Italy

Personal homepage

Download lecture slides: Characterization of Environment Properties based on Interactive Models
Download lecture slides: Probabilistic Graphical Models for Bayesian Learning of State Flow Models in Non-stationary Environments
Download lecture slides: Nonparametric Bayesian Activity Mining From Video

Biography:

Carlo S. Regazzoni (Ph.D.) is full professor of Cognitive Telecommunications Systems at DITEN, University of Genova, Italy. His main research interests include (see www.isip40.it) cognitive dynamic systems, adaptive and self-aware video processing, tracking and recognition, generative models and inference schemes based on hierarchical dynamic Bayesian networks, software and cognitive radio. He has been responsible of several national and EU funded research projects. He is currently coordinator of international PhD courses on Interactive and Cognitive Environments involving several European universities. He is author of peer-reviewed papers on international journals (90) and international conferences/books (350). He served as general chair (IEEE AVSS2009), technical program chair (IEEE ICIP2005, NSIP2002), associate editor (IEEE Trans on Image Processing, IEEE Trans on Mobile Computing, et al.), guest editor (Proceedings of the IEEE, IEEE Signal Processing Magazine et al.) in international conferences and Journals. He has served in many roles in governance bodies of IEEE SPS. He is currently serving as Vice President Conferences IEEE Signal Processing Society in 2015-2017.

Abstract:

Dynamic Bayesian Networks (DBN) can provide a coherent framework for representing, modeling and automatically managing at different abstraction levels complex interactive situations occurring in cognitive environments.

One of the key issues to design methods for learning DBNs is the estimation of multiple state flows models from dynamic observations. Such models can represent state dynamics under different stationarity conditions and their incremental online learning represent a necessary feature in different contexts, starting from Cognitive Dynamic Systems.

In this talk focus will be given to Bayesian learning techniques capable of learning generative switching models related to state flows stationary conditions that can represent implicitly or explicitly interactions between the represented dynamic pattern and the context in which the pattern operates.

Applications examples will be discussed including multitarget tracking, surveillance of transports, and, more in general, cognitive environments.


Powered by CONWIZ, © Copyright 2015 | Privacy Policy and Terms of Use | Page maintained by Jan Larsen