AESOP: AI Based Policy Learning for Predicting and Mitigating Network Service Impairments

Time: Thursday, December 14, 2017 - 2:00pm - 3:00pm
Type: Seminar Series
Presenter: Supratim Deb; Principal Inventive Scientist, AT&T Labs
Room/Office: Room 238
17 Hillhouse Avenue
New Haven, CT 06511
United States

Department of Electrical Engineering and the Yale Institute of Network Sciece (YINS) Seminar

Supratim Deb
Principal Inventive Scientist
AT&T Labs

"AESOP: AI Based Policy Learning for Predicting and Mitigating Network Service Impairments"

Abstract: In this talk, we present AESOP, a data-driven intelligent system to facilitate automatic learning of policies and rules for triggering remedial actions in Networks. AESOP combines best practices of operations' intelligence with variety of measurement data to learn and validate operational policies to mitigate service problems in networks. AESOP design addresses key challenges like learning from high-dimensional noisy data, capturing multiple fault models, very high service-cost of false positives, evolving network infrastructure (leading to changing data distribution), etc. We present the design of our system and show results from our ongoing experiments to show the operational efficiency that can be had from such policy driven automation.

Bio: Supratim Deb is currently a Principal Scientist with AT&T Labs. Previously, he worked at Bell Labs/Lucent as a DMTS, at MIT as a post-doc, and at Univ. of Illinois at Urbana-Champaign as PhD candidate. His current research interests are at the intersection of machine learning and networking. Over his career, he has designed several systems and products in the areas of data-driven intelligent systems, wireless and networked systems; he has also published numerous ACM/IEEE papers in the areas of applied machine learning, wireless systems, and networking.

Hosted by: Professor Leandros Tassiulas

Thursday, December 14, 2017
17 Hillhouse Room 328
2:00 PM