Skip to main content
Foundations of Information, Networks, and Decision Systems

Talk Information 11/14/2024

Title: Data-Driven Approach for Capacity Estimation and Polar Codes

Speaker: Haim Permuter
Date and Time: TBD 4:15PM ET
Location: Rhodes 310 and Zoom

Abstract: In this seminar, we introduce an innovative data-driven methodology for both capacity estimation and the design of polar codes, applicable to channels with and without memory. Our approach does not require explicit channel models, relying instead on observations of channel inputs and outputs. Key to our method is the development of a Neural Successive Cancellation (NSC) decoder, which integrates neural networks to replace core elements of the traditional Successive Cancellation (SC) decoder, such as the check-node, bit-node, and soft decision components. This NSC decoder, augmented by a neural network that embeds channel outputs into the SC decoder’s input space, ensures computational efficiency and consistency, independent of channel memory size.

Additionally, we present a principled framework for neural estimation and optimization of information measures. This includes the proposal of a novel Directed Information Neural Estimator (DINE), complementing the Mutual Information Neural Estimation (MINE). These estimators are optimized over continuous and discrete channel input spaces to accurately estimate feedforward and feedback capacities of general channels. While capacity estimation is the primary application, our techniques extend to scenarios such as probability density estimation for processes with memory, causality identification, and various machine learning applications.

Based on joint work with Ziv Aharoni, Ziv Goldfeld, Henry Pfister, Dor Tzur, and Bashar Hulihal.

Bio: Haim Permuter received his B.Sc. (summa cum laude) from Ben-Gurion University (BGU) and Ph.D. from Stanford University, both in Electrical Engineering, in 1997 and 2008, respectively. Between 1997-2004, he served as a scientific research officer in an R&D unit in the Israeli Defense Forces. In summer 2002 he worked for IBM, Almaden research center. He is a recipient of several rewards including Eshkol Fellowship, Wolf Award, Fulbright Fellowship, Stanford Graduate Fellowship, U.S.-Israel Binational Science Foundation Bergmann Memorial Award, and Allon Fellowship. Haim joined the faculty of Electrical Engineering Department at BGU in Oct 2008 as a tenure-track faculty, and is now a Professor, Luck-Hille Chair in Electrical Engineering. Haim also serves as head of the communication, cyber and information track in his department. Haim served on the editorial boards of the IEEE Transactions on Information Theory in 2013-2016 and serves again from Jan 2023 till now.