Distributed Economic Consensus: Federated Learning and Related Applications
Event details
| Date | 26.10.2023 |
| Hour | 15:30 › 16:30 |
| Speaker | Prof. Sriram Vishwanath - University of Texas in Austin hosted by Prof. Emre Telatar (LTHI) |
| Location | |
| Category | Conferences - Seminars |
| Event Language | English |
This seminar was given on Monday 9th as part of the IC colloquium. Sriram offers a more detailed and mathematical version for IPG seminar
Distributed systems that utilize incentive alignment together with economic consensus algorithms are of significant interest for many domains, including distributed machine learning/AI, distributed wireless spectrum sharing and distributed resource allocation for both physical and virtual resources.
In this talk, we will move past the blockchain hype by first explaining the nature of distributed economic consensus and its difference from traditional distributed systems consensus algorithms. Then, we will delve into the benefits and limitations of such consensus mechanisms. Through illustrative examples including federated learning, distributed wireless spectrum sharing and related problems, we will understand the benefits of distributed economic consensus over both centralized schemes and other traditional, non-economic consensus algorithms.
Bio:
Sriram Vishwanath received the B. Tech. degree in Electrical Engineering from the Indian Institute of Technology (IIT), Madras, India in 1998, the M.S. degree in Electrical Engineering from California Institute of Technology (Caltech , Pasadena USA in 1999, and the Ph.D. degree in Electrical Engineering from Stanford University, Stanford, CA USA in 2003. Over the past 20 years, he is a Professor in the Chandra Department of Electrical and Computer Engineering at The University of Texas at Austin.
Sriram’s research is in the domains of blockchain systems, machine learning/AI and information & coding theory. He has over 250 refereed research papers, and multiple research awards. He works across a diverse set of areas and specializes in bringing the gap between theory and practice. In particular, he has been involved in multiple spinouts from The University of Texas. Sriram also runs a non-profit incubator/accelerator for deep tech startups.
Distributed systems that utilize incentive alignment together with economic consensus algorithms are of significant interest for many domains, including distributed machine learning/AI, distributed wireless spectrum sharing and distributed resource allocation for both physical and virtual resources.
In this talk, we will move past the blockchain hype by first explaining the nature of distributed economic consensus and its difference from traditional distributed systems consensus algorithms. Then, we will delve into the benefits and limitations of such consensus mechanisms. Through illustrative examples including federated learning, distributed wireless spectrum sharing and related problems, we will understand the benefits of distributed economic consensus over both centralized schemes and other traditional, non-economic consensus algorithms.
Bio:
Sriram Vishwanath received the B. Tech. degree in Electrical Engineering from the Indian Institute of Technology (IIT), Madras, India in 1998, the M.S. degree in Electrical Engineering from California Institute of Technology (Caltech , Pasadena USA in 1999, and the Ph.D. degree in Electrical Engineering from Stanford University, Stanford, CA USA in 2003. Over the past 20 years, he is a Professor in the Chandra Department of Electrical and Computer Engineering at The University of Texas at Austin.
Sriram’s research is in the domains of blockchain systems, machine learning/AI and information & coding theory. He has over 250 refereed research papers, and multiple research awards. He works across a diverse set of areas and specializes in bringing the gap between theory and practice. In particular, he has been involved in multiple spinouts from The University of Texas. Sriram also runs a non-profit incubator/accelerator for deep tech startups.
Practical information
- Informed public
- Free
Contact
- Michael Gastpar