IEEE Signal Processing Society IIT Kanpur Presents
IEEE SPS Seminar Series
The lecture series runs in hybrid mode. Each talk is of around 45-60 minutes, followed by a Q&A of 10-15 minutes. All talks are recorded and uploaded to YouTube for later view (subjected to the speaker's consent).
To ask the speaker a question, please use the Q&A function in Zoom to type your questions. The moderator (Dr. Ketan Rajawat) will collect questions through the talk and ask the questions after the talk.
Upcoming Talks (all date/time are in IST)
Toni Heittola
Talk on: 22 August 2023
4:30 P.M.
Talk Title: Deep Learning methods for AudioAI
Abstract: The goal of audio content analysis is to extract information about content of audio signals. Research in this domain focuses on three types of audio: speech, music and everyday audio. While content analysis of speech and music are well-established research fields with many commercial applications, the content analysis of everyday audio is a relatively new research field. In recent years, content analysis of everyday audio has gained a lot of interest in both academia and industry research. This talk will introduce audio content analysis, and give the basics of the analysis methods, audio representations and deep neural networks. The main focus of the talk will be on methods specifically used for everyday audio. In addition to the basics, the talk will also introduce a few advanced topics and methods in everyday scene analysis.
Speaker Bio: Toni Heittola is a Postdoctoral Research Fellow at Tampere University, Finland. He received his PhD degree in 2021 from Tampere University. He is the author of 36 scientific publications on the topic of sound event detection and is an important contributor to the Detection and Classification of Acoustic Scenes and Events (DCASE) Challenge organization. Overall, he has authored more than 60 scientific publications in the field of audio content analysis, which have been cited over 6400 times. He is also a key contributor to open science in the field, authoring more than 15 public data sets, multiple evaluation toolboxes and baseline systems for the DCASE Challenge over the years. Together with Tuomas Virtanen and Annamaria Mesaros, he presented the tutorial on DCASE at the 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing.
Talk Title: Deployable Reinforcement Learning (DRL): Dealing with challenges of Sparse Rewards in Robotics via Heavy-Tailed Policy Gradient
Abstract: Recent advancements in Artificial Intelligence (AI), such as AlphaZero and ChatGPT, have significantly impacted various fields. Reinforcement learning (RL) plays a crucial role in these achievements. However, deploying RL in real-world applications, including robotics, finance, and healthcare, presents challenges such as efficient exploration, scalability, domain adaptation, and safety. One key aspect common to all these challenges in RL is the design of effective reward functions, which are often assumed to be known but remain elusive in practice. In this talk, we will discuss our recent results in addressing these challenges, specifically focusing on sparse rewards in robotic applications. While designing sparse rewards may seem easier, it introduces significant exploration challenges that make traditional algorithms inefficient. To tackle this, we propose heavy-tailed policy gradient algorithms, which provide a promising solution. We derive precise sample complexity bounds for the proposed algorithms and demonstrate their effectiveness in both simulators and real robots.
Speaker Bio: Amrit Singh Bedi is an assistant research professor/scientist in the Computer Science Department at the University of Maryland, College Park, MD, USA. He obtained his Ph.D. in Electrical Engineering from IIT Kanpur, Kanpur, India, in 2018. Following his doctoral studies, he worked as a Research Associate within the Computational and Information Sciences Directorate at the US Army Research Laboratory (ARL) in Adelphi, MD, USA, from 2019 to 2022. His research interests lie in artificial intelligence (AI) for autonomous systems, with specific emphasis on scalable & sample-efficient reinforcement learning algorithms and trustworthy AI through reliable AI-generated text detection. Currently, he is working on developing "Deployable Reinforcement Learning" algorithms with applications in robotics, finance, etc. His paper was selected as one of the Best Paper Finalists at the 2017 IEEE Asilomar Conference on Signals, Systems, and Computers. He received an honorable mention from the IEEE Robotics and Automation Letters in 2020. He was awarded the Amazon Research Award in 2022.
Previous Talks (Including Recordings) :
Talk on: 20th Nov. 2020, 4 pm
Praneeth is a senior researcher at Microsoft Research India, Bengaluru and an adjunct professor at TIFR, Mumbai. He is also a faculty associate of ICTS, Bengaluru. Prior to this, he was a postdoctoral researcher at Microsoft Research New England in Cambridge, MA. He obtained his MS and PhD in ECE from UT Austin. Before joining UT, he worked in the Quantitative Analysis group at Goldman Sachs, Bangalore. He obtained his B-Tech in Electrical Engineering from IIT Bombay.
Talk Title:
Projection Efficient Subgradient Method and Optimal Nonsmooth Frank-Wolfe Method . (Click for Details)
Talk on: 27th Nov. 2020
Sandeep is a Professor and Dean at the School of Technology and Computer Science in Tata Institute of Fundamental Research in Mumbai. He received his B.Tech. in Mechanical Engineering from IIT Delhi (1989) and his M. S. in Statistics and Ph.D. in Operations Research from Stanford University (1993). He then worked for a financial credit insurance company (American Credit Indemnity) in Baltimore, US (2 years) followed by an year long stint in Andersen Consulting in India. From December 1996 to December 2002, he was a faculty in the Operations Research Group in the Department of Mechanical Engineering at IIT Delhi. Since then he has been at TIFR. He has held visiting positions at many places including at Columbia University, Stanford University and Indian School of Business. He is currently on the editorial board of Stochastic Systems. Earlier he has been on editorial boards of Mathematics of Operations Research, Management Science and ACM TOMACS. He is a recipient of IBM faculty partnership award in the year 2001-02.
Talk Title:
Partition identification using multi-armed bandits for heavy-tailed distributions . (Click for Details)
Talk on: 11th Dec. 2020
Manjesh is an Assistant Professor at the centre for Industrial Engineering and Operations Research (IEOR), Indian Institute Technology Bombay, Mumbai, India, since January 2016. Before joining IITB, he was a postdoctoral researcher at the Department of ECE, Boston University, USA for two years. He obtained his Ph.D. degree from University of Avignon, France (and INRIA, Sophia Antipolis, France) in 2013 and a M.Sc (Engg) degree from Indian Institute of Science, Bangalore, India, in 2009. He did his bachelor's degree from National Institute of Technology Bhopal (MANIT), India in 2004. Between 2004-2007, he worked as Scientist-B at Centre for Artificial Intelligence and Robotics, Bangalore.
His research interests are broadly in Machine learning and Communication networks. In machine learning, his interest is in development and analysis of online learning algorithms that efficiently exploit the problem structure and work with limited feedback. In communication networks, he is interested in resource allocation problems and study of various economics issues related to the ongoing net-neutrality debate.
Talk on: 18th Dec. 2020
Sandeep Kumar is an assistant professor in the Department of Electrical Engineering at the Indian Institute of Technology Delhi. He is also an associated faculty member with the School of Artificial Intelligence and Bharti School of telecommunications Technology and Management, at IIT Delhi. He is a visiting research fellow at the Hong Kong University of Science and Technology and the Inspire fellow awarded by the Department of Science and Technology, Government of India. His current research focuses on large scale non-convex optimization, graphical models, and signal processing techniques for applications in data analytics, communications, and networks. The contributions from his research have been published in several reputed journals and conferences including IEEE TSP, JMLR, and NeurIPS. He received the M.Tech and Ph.D. degrees from the IIT Kanpur, in 2013 and 2017 respectively. Post which he completed a post-doctoral fellowship from the Hong Kong University of Science and Technology from 2017-2020.
Talk on: 22 Jan 2021
Rahul Vaze is a reader (Tenured, eq. to Assoc. Prof.) at the School of Technology and Computer Science in Tata Institute of Fundamental Research in Mumbai. He received his Bachelor of Engineering (B.E.) Electronics from Madhav Institute of Technology and Science (2002), Master of Engineering (M.E.) in Telecommunication from Indian Institute of Science, Bangalore India (2004) and Ph.D. from University of Texas at Austin (2009). He also worked as a Design Engineer with Beceem Communications Pvt. Ltd., Bangalore, where he was involved in the design and development of physical layer algorithms for the IEEE 802.16-e standard. His research interests are broadly in Resource Allocation, Online Algorithms, and Wireless Communication Networks.
Talk on: 12 Feb 2021
Prashanth L.A. is an Assistant Professor in the Department of Computer Science and Engineering at Indian Institute of Technology Madras. Prior to this, he was a postdoctoral researcher at the Institute for Systems Research, University of Maryland - College Park from 2015 to 2017 and at INRIA Lille - Team SequeL from 2012 to 2014. From 2002 to 2009, he was with Texas Instruments (India) Pvt Ltd, Bangalore, India.
He received his Masters and Ph.D degrees in Computer Science and Automation from Indian Institute of Science, in 2008 and 2013, respectively. He was awarded the third prize for his Ph.D. dissertation, by the IEEE Intelligent Transportation Systems Society (ITSS). He is the coauthor of a book entitled 'Stochastic Recursive Algorithms for Optimization: Simultaneous Perturbation Methods', published by Springer in 2013. His research interests are in reinforcement learning, simulation optimization and multi-armed bandits, with applications in transportation systems, wireless networks and recommendation systems.
Talk Title: Non-asymptotic bounds for stochastic optimization with a biased noisy gradient oracle (Click for Details)
Talk on: 26 Feb 2021
Himanshu Tyagi is an Assistant Professor at the Department of Electrical Communication Engineering and an associate Faculty at Robert Bosch Center for Cyber-Physical Systems at Indian Institute of Science. He received his Dual degree in Electrical Engineering from IIT Delhi (2007) and Ph.D. in Electronics and Communication Engineering from the University of Maryland (2013). Post which he was a Postdoc at the ITA Center, UCSD (2014). His research interests are Information theory, statistics, cryptography, machine learning, distributed intelligence systems, socio-technical systems.
Talk Title: Stochastic optimization of convex functions under communication constraints (Click for Details)
Talk on: 5 Mar 2021
Alec Koppel is a Research Scientist at the U.S. Army Research Laboratory in the Computational and Information Sciences Directorate since September of 2017. He completed his Master's degree in Statistics and Doctorate in Electrical and Systems Engineering, both at the University of Pennsylvania (Penn) in August of 2017. Before coming to Penn, he completed his Master's degree in Systems Science and Mathematics and Bachelor's Degree in Mathematics, both at Washington University in St. Louis (WashU), Missouri. He is a recipient of the 2016 UPenn ESE Dept. Award for Exceptional Service, an awardee of the Science, Mathematics, and Research for Transformation (SMART) Scholarship, a co-author of Best Paper Finalist at the 2017 IEEE Asilomar Conference on Signals, Systems, and Computers, and an awardee of the 2020 ARL Director's Research Initiative Translational Research Challenge. His research interests are in optimization and machine learning. Currently, he focuses on scalable Bayesian learning, reinforcement learning, and decentralized optimization, with an emphasis on problems arising in robotics and autonomy.
Organizers:
This event is organized by the Student chapter of IEEE Signal Processing Society at IIT Kanpur. The members include:
Faculty:
Current Student Members:
Parampreet Singh
Sumit Kumar
Bebina Hidangmayum
Kusum Rajpurohit
Kapil Singla
Previous Team:
Priyadarshini Dwivedi