Published Nov 30, 2023

Thomas Dietterich: From the Foundations

Thomas Dietterich delves into advancing machine learning with error-correcting codes, the importance of causal understanding in AI, and tackling the limitations of large language models, emphasizing innovative methods to enhance accuracy and systematic reasoning in AI systems.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Hugo Larochelle: Deep Learning as Science
Answers 383 questions
François Chollet: Keras and Measures of Intelligence
Answers 383 questions
Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science
Answers 383 questions
Zachary Lipton: Where Machine Learning Falls Short
Answers 383 questions
Soumith Chintala: PyTorch
Answers 383 questions
Nicholas Thompson: AI and Journalism
Answers 383 questions
Ted Underwood: Machine Learning and the Literary Imagination
Answers 383 questions
Some Changes at The Gradient
Answers 383 questions
Ed Grefenstette: Language, Semantics, Cohere
Answers 383 questions
Hello World from The Gradient Podcast!
Answers 383 questions
David Pfau: Manifold Factorization and AI for Science
Answers 383 questions
Martin Wattenberg: ML Visualization and Interpretability
Answers 383 questions
Been Kim: Interpretable Machine Learning
Answers 383 questions
Daniel Situnayake: AI on the Edge
Answers 383 questions
Sebastian Raschka: AI Education and Research
Answers 383 questions

Dexa/Daniel Bashir

Thomas Dietterich: From the Foundations

Topics covered

Popular Clips

Neural Networks and Error Correcting Codes

Causal Understanding and AI

Exploring Knowledge Level Learning

Ensembles and Variance

Rethinking LLMs

Challenging Off-the-Shelf Algorithms

Language and Reasoning

Large Language Models

Exogenous Variables

Solving Multiclass Learning Problems

Understanding Artificial Intelligence

Understanding Bias

Improving LLMs

Symbol Level Learning

Episode Highlights

Error-Correcting Codes

Causal Understanding

Large Language Models

Related Episodes

Hugo Larochelle: Deep Learning as Science

François Chollet: Keras and Measures of Intelligence

Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science

Zachary Lipton: Where Machine Learning Falls Short

Soumith Chintala: PyTorch

Nicholas Thompson: AI and Journalism

Ted Underwood: Machine Learning and the Literary Imagination

Some Changes at The Gradient

Ed Grefenstette: Language, Semantics, Cohere

Hello World from The Gradient Podcast!

David Pfau: Manifold Factorization and AI for Science

Martin Wattenberg: ML Visualization and Interpretability

Been Kim: Interpretable Machine Learning

Daniel Situnayake: AI on the Edge

Sebastian Raschka: AI Education and Research

Thomas Dietterich: From the Foundations

Topics covered

Popular Clips

Episode Highlights

Error-Correcting CodesThomas Dietterich explores the use of error-correcting output codes and ensemble methods to enhance machine learning models. These techniques address multiclass classification challenges and improve model robustness and accuracy.

Error-Correcting Codes

Causal UnderstandingThomas Dietterich explores the boundaries of statistical learning, emphasizing the need for causal reasoning to enhance AI systems. He discusses the significance of Judea Pearl's causal theories and the importance of testing for systematic understanding in machine learning.

Causal Understanding

Large Language Models

Related Episodes