Episode 15: Martín Arjovsky, INRIA, on benchmarks for robustness and geometric information theory

1

Introduction

00:00 • 2min

2

How Did You Become Interested in Machine Learning?

01:46 • 4min

3

How Machine Learning Can Reduce the Compression of Bidio Compressors?

05:29 • 2min

4

The Last Chain in the Story, Vasically

07:56 • 3min

5

Is There a Deadline for a Clear in Six Weeks?

11:08 • 2min

6

How to Make a Machine Learning Machine

13:22 • 2min

7

The Invaliant Causal Prediction Paper

15:17 • 2min

8

Is There a Gray in the Sand?

17:13 • 4min

9

The Distribution of Locations Is Not the Same

21:18 • 2min

10

Using a Dataset to Find Markers for Canser in Space

23:15 • 4min

11

We Need More Benchmarks on Robastness

27:13 • 4min

12

Is There a Benchmark for Out of Solution Gennization?

31:27 • 2min

13

How to Get a 100 Percent Curacy in a Linear Problem?

33:14 • 2min

14

Then You Throw Away Data Points

35:16 • 4min

15

What Makes a Bench Mark Take Off?

39:18 • 2min

16

How to Carry a Good Benchmark?

41:05 • 5min

17

Is the Entropy of a Negative Infinite?

45:35 • 3min

18

What's the Half That's Missing?

48:15 • 4min

19

Nour Networks Are Compromising With Entropy

52:02 • 4min

20

High Blood Pressure - Cause and Effect

55:57 • 4min

21

Papers That Go From Intuition to Solid Toy Tasks

59:55 • 2min

22

The Best Papers From Shorty Ofr So Many

01:01:31 • 3min

23

The Uncertainty in Prediction

01:04:33 • 2min

24

I Think Scale Is Not the Answer

01:06:33 • 4min

25

Is There a Difference Between Ganis and Unsupervised Learning?

01:10:41 • 2min

26

How to Make a Product Work for You

01:12:38 • 2min

27

How to Learn the Rail Without a Value Function?

01:14:38 • 2min

28

The Best Researches That You've Worked With?

01:16:48 • 2min

29

Working on a Porn, Is It Important?

01:18:46 • 4min

30

Exploration in Infant Psychology

01:22:30 • 2min

31

The Importance of Uncertainty Estimation and Nanomalyetaction

01:24:17 • 2min