Machine Learning Street Talk (MLST)6 Des 2020

#032- Simon Kornblith / GoogleAI - SimCLR and Paper Haul!

This week Dr. Tim Scarfe, Sayak Paul and Yannic Kilcher speak with Dr. Simon Kornblith from Google Brain (Ph.D from MIT). Simon is trying to understand how neural nets do what they do. Simon was the second author on the seminal Google AI SimCLR paper. We also cover "Do Wide and Deep Networks learn the same things?", "Whats in a Loss function for Image Classification?", and "Big Self-supervised models are strong semi-supervised learners". Simon used to be a neuroscientist and also gives us the story of his unique journey into ML.

00:00:00 Show Teaser / or "short version"

00:18:34 Show intro

00:22:11 Relationship between neuroscience and machine learning

00:29:28 Similarity analysis and evolution of representations in Neural Networks

00:39:55 Expressability of NNs

00:42:33 Whats in a loss function for image classification

00:46:52 Loss function implications for transfer learning

00:50:44 SimCLR paper

01:00:19 Contrast SimCLR to BYOL

01:01:43 Data augmentation

01:06:35 Universality of image representations

01:09:25 Universality of augmentations

01:23:04 GPT-3

01:25:09 GANs for data augmentation??

01:26:50 Julia language

@skornblith

https://www.linkedin.com/in/simon-kornblith-54b2033a/

https://arxiv.org/abs/2010.15327

Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth

https://arxiv.org/abs/2010.16402

What's in a Loss Function for Image Classification?

https://arxiv.org/abs/2002.05709

A Simple Framework for Contrastive Learning of Visual Representations

https://arxiv.org/abs/2006.10029

Big Self-Supervised Models are Strong Semi-Supervised Learners

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(271)

What Most People Get Wrong About Evolution | Akarsh Kumar

A lot of people in AI treat evolution as a dumb fallback, basically random search for when you can't take a gradient. Akarsh Kumar thinks that is wrong. Selection hangs on to partial solutions, so mut...

10 Okt 47min

How a Voice Agent Learns the Rhythm of Conversation — Shawn Wen

Tsung-Hsien (Shawn) Wen, CTO of PolyAI, tells Tim Scarfe why voice agents are harder than text agents. Voice adds time, and a good conversation depends on adapting to the person on the line, not just ...

1 Okt 1h 10min

Who Checks a Proof No Human Can Read? — Leo de Moura

Leonardo de Moura created Lean and co-created Z3. ---This episode is sponsored by Parallel.Parallel, where agents find answers: web search, extraction and deep research APIs built for AI agents.Start ...

30 Sep 1h 14min

When AI Research Starts Moving Faster Than Human Research - Zhengyao Jiang

Weco let an AI coding agent rewrite the harness around another agent for eight days: its code, prompts and tools, while the underlying language model stayed fixed. Tim Scarfe asks Weco co-founder Zhen...

26 Sep 43min

How Deep Learning Finally Cracked Messy Tables - Frank Hutter

Frank Hutter, co-founder of Prior Labs, talks about TabPFN, a tabular foundation model that makes predictions in a single forward pass, and the research behind it.TabPFN is pre-trained on synthetic da...

23 Sep 1h 53min

Why Scaling Prediction Cannot Create Intelligence - Alexander Mattick

Alexander Mattick is a researcher at Fraunhofer IIS and a PhD researcher at the University of Technology Nuremberg (UTN), and a regular on Yannic Kilcher's Discord. He first came on MLST in 2022, afte...

21 Sep 2h 14min

How Physical AI Learns Across Language, Video and Action — Ming-Yu Liu

The car making a left turn at the start of this episode was never filmed. Cosmos 3 generated it. Ming-Yu Liu, who leads the Cosmos research at NVIDIA, explains how one model can describe a video, gene...

15 Sep 25min

Speech Recognition Is Not a Solved Problem — Pavan Kumar Reddy

Pavan Kumar Reddy leads audio research at Mistral AI. He joins Tim Scarfe for a deep technical tour of Voxtral — and explains why the frontier of deployed voice is still a cascade of specialised model...

14 Sep 1h 42min