doi.bio/yarin_gal


Yarin Gal

Yarin Gal is an Associate Professor of Machine Learning at the Computer Science department at the University of Oxford. He is also a Tutorial Fellow in Computer Science at Christ Church, Oxford, a Turing AI Fellow at the Alan Turing Institute, and Director of Research at the UK Government's AI Safety Institute (AISI).

Education

Yarin Gal obtained his PhD from the Cambridge Machine Learning Group, working with Zoubin Ghahramani and funded by the Google Europe Doctoral Fellowship. Prior to that, he studied Computer Science at Oxford for a Master's degree under the supervision of Phil Blunsom.

Career

Before taking up his position at Oxford, Gal was a Research Fellow in Computer Science at St Catharine's College, Cambridge. He has also taught Advanced Machine Learning (2018-2019), Advanced Topics in Machine Learning (2021-2022), and Uncertainty in Deep Learning (2023-2024). In addition, he has taught machine learning at the NASA Frontier Development Lab, helping NASA make use of AI for the space program.

Research

Gal's research interests lie in the fields of linguistics, applied maths, and computer science, particularly the problems found at the intersections of these fields. His current research focuses on developing Bayesian techniques for deep learning, with applications in reinforcement learning. He has also worked on Bayesian modelling, approximate inference, natural language processing, Bayesian nonparametrics, Gaussian processes, inference algorithms for big data, and machine translation.

Gal has achieved notable research accomplishments in Bayesian statistics, approximate Bayesian inference, and Bayesian deep learning. He has also worked on applications such as computer vision, AI safety, and ML interpretability.

Publications

Yarin Gal has published extensively in the fields of machine learning, artificial intelligence, probability theory, and statistics. His publications include:

Youtube Videos

Youtube Title: Yarin Gal - Uncertainty in Deep Learning | MLSS Kraków 2023

Youtube Link: link

Youtube Channel Name: ML in PL

Youtube Channel Link: https://www.youtube.com/@MLinPLAssociation

Yarin Gal - Uncertainty in Deep Learning | MLSS Kraków 2023

Youtube Title: Yarin Gal -. Bayesian Deep Learning

Youtube Link: link

Youtube Channel Name: SMILES - Summer School of Machine Learning at SK

Youtube Channel Link: https://www.youtube.com/@smiles-summerschoolofmachi5505

Yarin Gal -. Bayesian Deep Learning

Youtube Title: Yarin Gal - Bayesian Deep Learning Pt.2

Youtube Link: link

Youtube Channel Name: SMILES - Summer School of Machine Learning at SK

Youtube Channel Link: https://www.youtube.com/@smiles-summerschoolofmachi5505

Yarin Gal - Bayesian Deep Learning Pt.2

Youtube Title: Exploring foundation models - Session 3

Youtube Link: link

Youtube Channel Name: The Alan Turing Institute

Youtube Channel Link: https://www.youtube.com/@TheAlanTuringInstituteUK

Exploring foundation models - Session 3

Youtube Title: Lightning lectures: The art of AI extrapolation | The Royal Society

Youtube Link: link

Youtube Channel Name: The Royal Society

Youtube Channel Link: https://www.youtube.com/@royalsociety

Lightning lectures: The art of AI extrapolation | The Royal Society

Youtube Title: Professor Yarin Gal's Keynote on Human-in-the-loop Bayesian Deep Learning (UNSURE 2020)

Youtube Link: link

Youtube Channel Name: UNSURE Workshop

Youtube Channel Link: https://www.youtube.com/@unsureworkshop8093

Professor Yarin Gal's Keynote on Human-in-the-loop Bayesian Deep Learning (UNSURE 2020)

Youtube Title: ESGW: Artificial Intelligence for Space - Panel Discussion

Youtube Link: link

Youtube Channel Name: spacegeneration

Youtube Channel Link: https://www.youtube.com/@spacegeneration

ESGW: Artificial Intelligence for Space - Panel Discussion

Youtube Title: MIC 2018 - Targeted Dropout and Bitrot

Youtube Link: link

Youtube Channel Name: MIC Media

Youtube Channel Link: https://www.youtube.com/@micmedia5591

MIC 2018 - Targeted Dropout and Bitrot

Youtube Title: Understanding Approximate Inference in Bayesian Neural Networks: A Joint Talk

Youtube Link: link

Youtube Channel Name: OATML research group

Youtube Channel Link: https://www.youtube.com/@oatmlresearchgroup9874

Understanding Approximate Inference in Bayesian Neural Networks: A Joint Talk

Youtube Title: BILLA SONIPAT ALA : Yaaran Gail (Official Video) Guri Nimana | Haryanvi Songs Harayanvi 2022

Youtube Link: link

Youtube Channel Name: White Hill Dhaakad

Youtube Channel Link: https://www.youtube.com/@WhiteHillDhaakad

BILLA SONIPAT ALA : Yaaran Gail (Official Video) Guri Nimana | Haryanvi Songs Harayanvi 2022

Youtube Title: Model Uncertainty in Deep Learning | Lecture 80 (Part 4) | Applied Deep Learning

Youtube Link: link

Youtube Channel Name: Maziar Raissi

Youtube Channel Link: https://www.youtube.com/@maziarraissi3569

Model Uncertainty in Deep Learning | Lecture 80 (Part 4) | Applied Deep Learning

Youtube Title: Evaluating Online Bayesian Inference in Sample-Based Approximate BNNs

Youtube Link: link

Youtube Channel Name: Updatable Machine Learning workshop @ ICML2022

Youtube Channel Link: https://www.youtube.com/@updatablemachinelearningwo9964

Evaluating Online Bayesian Inference in Sample-Based Approximate BNNs

Youtube Title: Trustworthy AI: Bayesian deep learning | AI FOR GOOD DISCOVERY

Youtube Link: link

Youtube Channel Name: AI for Good

Youtube Channel Link: https://www.youtube.com/@AIforGood

Trustworthy AI: Bayesian deep learning | AI FOR GOOD DISCOVERY

Youtube Title: Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics

Youtube Link: link

Youtube Channel Name: Alex Kendall

Youtube Channel Link: https://www.youtube.com/@AlexKendallNZ

Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics

Youtube Title: [ZC5] Toward a Tractable Solution for Human in the loop Reinforcement Learning

Youtube Link: link

Youtube Channel Name: SNU ECE BK21

Youtube Channel Link: https://www.youtube.com/@snuecebk2135

![ZC5] Toward a Tractable Solution for Human in the loop Reinforcement Learning