IEEE CSS Bangalore Chapter AGM on Feb 5th, 7 pm
We will have 2022 AGM (in virtual-mode) of IEEE Bangalore CSS Chapter on 5th February 2022 (Saturday) during 7-7:30pm. As the office-bearer members of the current Committee - Chair, Vice-chair, Secretary, Treasurer have served in these roles for ~1.5 years now, the Society will benefit with other members assuming these roles for 2022-23.
Hence, we request and encourage members to come forward and nominate other members of the Society (even self-nominations are welcome) for one of these roles.
Please send in your nominations by email privately in-confidence to the existing Chair (Prof Radhakant Padhi, Email: padhi@iisc.ac.in) and Secretary (Prof Sachit Rao, Email: sachit@iiitb.ac.in) providing the following information latest by 3-Feb-2022 Thursday ~5pm.
1) Name, affiliation, email address, and IEEE membership no. of the Nominator
2) Name, affiliation, email address, and IEEE membership no. of the Nominated person (self-nomination also welcome)
3) Nominated for which office-bearer position?
4) No. of years experience (industrial/academic/both) of the nominated person
5) No. of years of IEEE & IEEE-CSS memberships
6) A brief profile of the nominated person
Along with AGM, on the same date, we also plan to have two technical talks. The details of the first talk, which starts at 6 pm on Feb 5th, follow.
The second will be confirmed soon.
Regards
Sachit Rao
Secretary, IEEE Bangalore CSS Chapter
Date and Time
Location
Hosts
Registration
- Date: 05 Feb 2022
- Time: 06:00 PM to 10:00 PM
- All times are (GMT+05:30) Asia/Calcutta
- Add Event to Calendar
- Bangalore, Karnataka
- India
Speakers
Dr Anjaly Parayil of Microsoft Research
Towards Persistence: Policy search in Continuous Action Domains
Reinforcement learning is a framework for interactive decision-making with incentives sequentially revealed across time without a system model. Here I focus on policy search in continuous action space using stochastic policy gradient (PG) updates. In contrast to discrete action space, the non-convexity in continuous space poses a pathological challenge due to its limited convergence guarantees.
Therefore, in this talk, I will focus on our recent research efforts at the US Army Laboratory to address this gap. Inspired by persistent exploration and its tendency to avoid spurious behavior, we step towards persistent exploration in continuous space through policy parameterizations defined by distributions of heavier tails. Doing so invalidates smoothness conditions of the score function common to PG. Thus, the research establishes convergence rate to stationarity with the policy’s tail index and an exploration tolerance parameter introduced here for the first time. Further, we characterize the dependence of the set of local maxima on the tail index through metastability analysis of a suitably defined Markov chain, which identifies the convergence of heavy-tailed policies to wider peaks. This phenomenon yields improved stability to perturbations in supervised learning, which manifests in the improved performance of policy search, particularly when myopic and farsighted incentives are misaligned. The last part of the talk discusses real-world applications and future research directions.
Biography:
Anjaly Parayil is a researcher at Microsoft Research India. Prior to joining Microsoft, she was a postdoctoral research fellow with the US Army Research Laboratory in the Computational and Information Sciences Directorate Adelphi from September 2019 to August 2021. Her research interests include Multi-agent Control and Coordination, Reinforcement Learning, Bifurcation Theory, Optimization, State Estimation, Machine Learning Applied to Large-scale Datasets, and Bayesian Approaches. She received her Ph.D. in Aerospace Engineering from the Indian Institute of Science (IISc), India, in 2020.
Email:
Address:Microsoft Research, , Bangalore, Karnataka, India
Agenda
Tentative Agenda for the AGM:
1. Welcome address by Chair
2. 2021 Activities and Finance Report by the Secretary and Treasurer
3. Presentation of slate and Election of 2022 office-bearers and Execom
4. 2022 Plans: Activities & budget overview
5. Any ideas, suggestions & activity proposals from members
6. Any other matters (with Chair's permission)
7. Vote of Thanks