Past Events

Improved Strong Upper Bounds for Policy Iteration

Speaker: 

Shivaram Kalyanakrishnan

Organisers: 

Hariharan Narayanan

Time: 

Tuesday, 28 August 2018, 14:00 to 15:00

Venue: 

Abstract: Markov Decision Problems (MDPs) are a well-studied abstraction of sequential decision making. Policy Iteration (PI) is a classical, widely-used family of algorithms to compute an optimal policy for a given MDP.

Quadratic Loss Minimization with Portfolio and Intertemporal Wealth Constraints

Speaker: 

Andrew Heunis

Organisers: 

Sandeep K Juneja

Time: 

Tuesday, 24 July 2018, 16:00 to 17:00

Venue: 

We address a problem of stochastic optimal control motivated by portfolio optimization in mathematical finance, the goal of which is to minimize the expected value of a general quadratic loss function of the wealth at close of trade when there is

On Randomization and Combinatorics in Computational Geometry, Discrete Mathematics, and Combinatorial Representation Theory.

Speaker: 

Kunal Dutta

Organisers: 

Hariharan Narayanan

Time: 

Friday, 20 July 2018, 14:00 to 15:00

Venue: 

Abstract: In this talk we shall see three very different areas of applications of combinatorics in mathematics and computer science, illustrating different flavours of combinatorial reasoning.

Edit Distance Codes via Synchronization Strings

Speaker: 

Madhu Sudan

Organisers: 

Prahladh Harsha

Time: 

Wednesday, 11 July 2018, 14:00 to 15:00

Venue: 

I will describe a recent approach to designing codes that correct for "editing" errors - i.e., where an adversary is allowed to delete some of the symbols in a string being transmitted and insert new symbols. The classical Hamming model of errors

Pages