Panelist: Prof. Vahid Tarokh
Title: FAST Lexicographic Agile Perception (FLAP) Integrates Decision and Control in a Spike Resolved Sensorimotor Program
Abstract: In this presentation the speaker discusses some of their activities in deriving neurally inspired principles for fast decision-making where task-relevant information is poor, occluded or distracted. The work is inspired by the remarkable capabilities, yet amazingly compact neural systems of insects. The emphasis of this work is on understanding strategies for parsing complex sensory scenes with multiple modalities, adaptive attention, and lexicographic representations using the idea that percpetion is inherently coupled to actions. The plan is to translate these ideas to high-fidelity simulations and robotic platforms to address the needs of autonomy and decision-making on rapid timescales.
This presentation is based on joint papers with Simon Sponberg & Hannah Choi (Georgia Tech), Jeff Riffell (UW-Seattle), Jamie Theobald (FIU), and Silvia Ferrari (Cornell) based on the work supported by AFOSR Grant No. FA9550-22-1-0315.
Bio-Overview: Prof. Vahid Tarokh worked at AT&T Labs-Research until August 2000. In September 2000, he joined the Massachusetts Institute of Technology (MIT) as an Associate Professor of Electrical Engineering and Computer Science. In June 2002, he joined Harvard University as a Gordon McKay Professor of Electrical Engineering and Hammond Vinton Hayes Senior Research Fellow. Prof. Tarokh was named Perkins Professor of Applied Mathematics in 2005. In Jan 2018, he joined Duke University, as the Rhodes Family Professor of Electrical and Computer Engineering, Bass Connections Endowed Professor, and Professor of Computer Science, and Mathematics. From Jan 2018 to May 2018, he was a Gordon Moore Distinguished Scholar at the California Institute of Technology (CALTECH). During Jan 2019-Dec 2022, he served as a Microsoft Data Science Investigator at Duke University.