Seminars Sorted by Series

We consider sequential prediction with expert advice when the data are generated stochastically, but the distributions generating the data may vary arbitrarily among some constraint set. We quantify relaxations of the classical I.I.D. assumption in...

In this talk I will introduce our next generation of graph neural networks. GNNs have the property that they are invariant to permutations of the nodes in the graph and to rotations of the graph as a whole. We claim this is unnecessarily restrictive...

Some of the aspects of the world around us are captured in natural language and refer to semantic high-level variables, which often have a causal role (referring to agents, objects, and actions or intentions). These high-level variables also seem to...

I will introduce Generalized Energy Based Models (GEBM) for generative modelling. These models combine two trained components: a base distribution (generally an implicit model), which can learn the support of data with low intrinsic dimension in a...

For autonomous robots to operate in the open, dynamically changing world, they will need to be able to learn a robust set of skills from relatively little experience. This talk begins by introducing Grounded Simulation Learning as a way to bridge...

Unsupervised learning, in particular learning general nonlinear representations, is one of the deepest problems in machine learning. Estimating latent quantities in a generative model provides a principled framework, and has been successfully used...

In handling wide range of experiences ranging from data instances, knowledge, constraints, to rewards, adversaries, and lifelong interplay in an ever-growing spectrum of tasks, contemporary ML/AI research has resulted in thousands of models...

Probably Approximately Correct (PAC) learning has attempted to analyse the generalisation of learning systems within the statistical learning framework. It has been referred to as a ‘worst case’ analysis, but the tools have been extended to analyse...

There are three core orthogonal problems in reinforcement learning: (1) Crediting actions (2) generalizing across rich observations (3) Exploring to discover the information necessary for learning. Good solutions to pairs of these problems are...

A brief review will be provided first on how deep learning has disrupted speech recognition and language processing industries since 2009. Then connections will be drawn between the techniques (deep learning or otherwise) for modeling speech and...

Suppose you are monitoring discrete events in real time. Can you predict what events will happen in the future, and when? Can you fill in past events that you may have missed? A probability model that supports such reasoning is the neural Hawkes...

Classical algorithms typically provide "one size fits all" performance, and do not leverage properties or patterns in their inputs. A recent line of work aims to address this issue by developing algorithms that use machine learning predictions to...

Many challenging problems in modern applications amount to finding relevant results from an enormous output space of potential candidates, for example, finding the best matching product from a large catalog or suggesting related search phrases on a...

Please Note: The seminars are not open to the general public, but only to active researchers.

Register here for this event: https://docs.google.com/forms/d/e/1FAIpQLScJ-BUVgJod6NGrreI26pedg8wGEyPhh3WMDskE1hIac_Yp3Q/viewform

The affinity between...

Please Note: The seminars are not open to the general public, but only to active researchers.

Register here for this event: https://docs.google.com/forms/d/e/1FAIpQLScJ-BUVgJod6NGrreI26pedg8wGEyPhh3WMDskE1hIac_Yp3Q/viewform

What interplay of dynamics...

Cubic fourfolds behave in many ways like K3 surfaces. Certain cubics - conjecturally, the ones that are rational - have specific K3s associated to them geometrically. Hassett has studied cubics with K3s associated to them at the level of Hodge...

In this talk I will focus on two examples of K3 categories: bounded derived categories of (twisted) coherent sheaves and K3 categories associated with smooth cubic fourfolds. The group of autoequivalences of the former has been intensively studied...

Esnault asked whether a smooth complex projective variety with infinite fundamental group has a nonzero symmetric differential, meaning a section of some symmetric power of the cotangent bundle. We prove a partial result in this direction, using...

The relevant preprints are: arXiv:1405.5154 "The Fano variety of lines and rationality problem for a cubic hypersurface", Sergey Galkin, Evgeny Shinder arXiv:1405.4902 "On two rationality conjectures for cubic fourfolds", Nicolas Addington

I will outline a construction of "tropical current", a positive closed current associated to a tropical variety. I will state basic properties of tropical currents, and discuss how tropical currents are related to a version of Hodge conjecture for...

I will discuss some of the topology of the fibers of proper toric maps and a combinatorial invariant that comes out of this picture. Joint with Luca Migliorini and Mircea Mustata.

Report on R. Virk's arXiv:1406.4855v3. This is a fun, short and simple note with variations on the well-known theme by G. Laumon that the Euler characteristics with and without compact supports coincide.

Beauville and Voisin proved that decomposable cycles (intersections of divisors) on a projective K3 surface span a 1-dimensional subspace of the (infinite-dimensional) group of 0-cycles modulo rational equivalence. I will address the following...

I will explain how infinite sequences of flops give rise to some interesting phenomena: first, an infinite set of smooth projective varieties that have equivalent derived categories but are not isomorphic; second, a pseudoeffective divisor for which...

What are the possible Hodge numbers of a smooth complex projective variety? We construct enough varieties to show that many of the Hodge numbers can take all possible values satisfying the constraints given by Hodge theory. For example, there are...