## Time:

## Venue:

Auto-bidding is a framework of ad auctions where every advertiser can tell their long-term goals, such as budget, target return on spend (RoS), etc., to an auto-bidding agent interface.

Speaker:

Agniv Bandyopadhyay, TIFR

Friday, 24 March 2023, 14:30 to 15:30

Auto-bidding is a framework of ad auctions where every advertiser can tell their long-term goals, such as budget, target return on spend (RoS), etc., to an auto-bidding agent interface.

Speaker:

Santanu Das, TIFR

Friday, 17 March 2023, 16:00 to 17:30

We examine gradient descent on unregularised logistic regression problems, with homogeneous linear predictors on linearly separable datasets. We show the predictor converges to the direction of the max-margin (hard margin SVM) solution.

Mihir Vahanwala

Wednesday, 5 April 2023, 16:00 to 17:00

The Skolem, Positivity, and Ultimate Positivity problems for Linear Recurrence Sequences (LRS) are number-theoretic problems whose decidability has been open for decades.

Speaker:

Varun Ramanathan, TIFR

Friday, 10 March 2023, 16:00 to 17:00

I will complete a couple of proofs from last week's student seminar on determinantal complexity. I will recall the required background for the proofs. On the way, we will learn a cute linear-algebraic fact.

Speaker:

Varun Ramanathan, TIFR

Friday, 3 March 2023, 16:30 to 17:30

We will introduce the notion of determinantal complexity, one of the main characters in the VP vs VNP question, which is the algebraic analogue of the P vs NP question.

Debraj Chakraborty

Tuesday, 21 March 2023, 16:00 to 17:00

We show how to combine techniques from formal methods and learning for online computation of a strategy that aims at optimizing the expected long-term reward in large systems modelled as Markov decision processes.

Speaker:

Friday, 24 February 2023, 16:00 to 17:00

Fair division of a set of resources among several agents is a commonly occurring problem in many real-world settings.

Speaker:

Eeshan Modak, TIFR

Friday, 17 February 2023, 16:00 to 17:00

In the composite hypothesis testing setting, the detector receives n i.i.d. samples either from a distribution p∈P or from a distribution q∈Q. It then decides the correct set from which the samples were drawn.

Speaker:

Malhar Ajit Managoli, TIFR

Friday, 10 February 2023, 16:00 to 17:00

Fermat's two square theorem states that:

An odd prime p can be written as a sum of two squares if and only if p = 1 (mod 4)

Sabyasachi Chatterjee

Friday, 10 February 2023, 14:00 to 15:00

We formulate a general cross validation framework for signal denoising. The general framework is then applied to nonparametric regression methods such as Trend Filtering and Dyadic CART.