Doorkeeper

[AIP Seminar] Talk by Prof. Shai Ben-David (University of Waterloo/Vector Institute, Canada) on "Learning probability distributions; what can, what can't be done."

Mon, 11 Mar 2024 10:30 - 12:00 JST
Online Link visible to participants
Register
Free admission
Registration closes 11 Mar 12:00
-Passcode: jPcVrt49uK -Time Zone:JST -The seats are available on a first-come-first-served basis. -When the seats are fully booked, we may stop accepting applications. -Simultaneous interpretation will not be available.
There is room for 205 more people

Description

Date and Time:
March 11, 2024: 10:30 am - 12:00 am (JST)
Venue: Online and Open Space at the RIKEN AIP Nihonbashi office*
*The Open Space; AIP researchers are only available.

TITLE: Learning probability distributions; what can, what can't be done.

SPEAKER: Prof. Shai Ben-David (University of Waterloo/Vector Institute, Canada)

ABSTRACT:
A possible high-level description of statistical learning is that it aims to learn about some unknown probability distribution ("environment”) from samples it generates ("training data”). In its most general form, assuming no prior knowledge and asking to find accurate approximations to the data generating distributions (a.k.a. density estimation), there can be no success guarantee. In this talk I will discuss two major directions of relaxing that too hard problem.

First, I will address the situation under common prior knowledge assumption - I will describe settling the question of the sample complexity of learning mixtures of Gaussians.

I will also mention unpublished recent results about characterization of the learnable families of distributions.

Secondly, I will address what can be learnt about unknown distributions when no prior knowledge is applied. I will describe a surprising result. Namely, the independence from set theory of a basic statistical learnability problem. As a corollary, I will show that there can be no combinatorial dimension that characterizes the families of random variables that can be reliably learnt (in contrast with the known VC-dimension-like characterizations of common supervised learning tasks).

Both parts of the talks use novel notions of sample compression schemes as key components.

The first part is based on joint work with Hasan Ashiani, Nick Harvey, Chris Law, Abas Merhabian and Yaniv Plan and the second part on work with Shay Moran, Pavel Hrubes, Amir Shpilka and Amir Yehudayoff. The recent characterization results are with my student Tosca lechner.


All participants are required to agree with the AIP Seminar Series Code of Conduct.
Please see the URL below.
https://aip.riken.jp/event-list/termsofparticipation/?lang=en

RIKEN AIP will expect adherence to this code throughout the event. We expect cooperation from all participants to help ensure a safe environment for everybody.


About this community

RIKEN AIP Public

RIKEN AIP Public

Public events of RIKEN Center for Advanced Intelligence Project (AIP)

Join community