b'Yarin Gal'

research

∙ 08/25/2023

Fine-tuning can cripple your foundation model; preserving features may be the solution

Pre-trained foundation models, owing primarily to their enormous capacit...

0 Jishnu Mukhoti, et al. ∙

research

∙ 07/23/2023

In-Context Learning in Large Language Models Learns Label Relationships but Is Not Conventional Learning

The performance of Large Language Models (LLMs) on downstream tasks ofte...

0 Jannik Kossen, et al. ∙

research

∙ 07/20/2023

LLM Censorship: A Machine Learning Challenge or a Computer Security Problem?

Large language models (LLMs) have exhibited impressive capabilities in c...

0 David Glukhov, et al. ∙

research

∙ 06/26/2023

BatchGFN: Generative Flow Networks for Batch Active Learning

We introduce BatchGFN – a novel approach for pool-based active learning ...

0 Shreshth A. Malik, et al. ∙

research

∙ 06/02/2023

ReLU to the Rescue: Improve Your On-Policy Actor-Critic with Positive Advantages

In this paper, we introduce a novel method for enhancing the effectivene...

0 Andrew Jesson, et al. ∙

research

∙ 05/27/2023

The Curse of Recursion: Training on Generated Data Makes Models Forget

Stable Diffusion revolutionised image creation from descriptive text. GP...

0 Ilia Shumailov, et al. ∙

research

∙ 04/17/2023

Prediction-Oriented Bayesian Active Learning

Information-theoretic approaches to active learning have traditionally f...

0 Freddie Bickford Smith, et al. ∙

research

∙ 04/07/2023

Revisiting Automated Prompting: Are We Actually Doing Better?

Current literature demonstrates that Large Language Models (LLMs) are gr...

0 Yulin Zhou, et al. ∙

research

∙ 02/21/2023

Differentiable Multi-Target Causal Bayesian Experimental Design

We introduce a gradient-based approach for the problem of Bayesian optim...

0 Yashas Annadani, et al. ∙

research

∙ 02/19/2023

Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation

We introduce a method to measure uncertainty in large language models. F...

0 Lorenz Kuhn, et al. ∙

research

∙ 12/28/2022

On Pathologies in KL-Regularized Reinforcement Learning from Expert Demonstrations

KL-regularized reinforcement learning from expert demonstrations has pro...

0 Tim G. J. Rudner, et al. ∙

research

∙ 12/15/2022

CLAM: Selective Clarification for Ambiguous Questions with Large Language Models

State-of-the-art language models are often accurate on many question-ans...

0 Lorenz Kuhn, et al. ∙

research

∙ 11/23/2022

Benchmarking Bayesian Deep Learning on Diabetic Retinopathy Detection Tasks

Bayesian deep learning seeks to equip deep neural networks with the abil...

0 Neil Band, et al. ∙

research

∙ 11/13/2022

Discovering Long-period Exoplanets using Deep Learning with Citizen Science Labels

Automated planetary transit detection has become vital to prioritize can...

0 Shreshth A. Malik, et al. ∙

research

∙ 09/27/2022

Exploring Low Rank Training of Deep Neural Networks

Training deep neural networks in low rank, i.e. with factorised layers, ...

0 Siddhartha Rao Kamalakara, et al. ∙

research

∙ 08/19/2022

Exploring the Limits of Synthetic Creation of Solar EUV Images via Image-to-Image Translation

The Solar Dynamics Observatory (SDO), a NASA multi-spectral decade-long ...

15 Valentina Salvatelli, et al. ∙

research

∙ 08/01/2022

Unifying Approaches in Data Subset Selection via Fisher Information and Information-Theoretic Quantities

The mutual information between predictions and model parameters – also r...

4 Andreas Kirsch, et al. ∙

research

∙ 06/14/2022

Prioritized Training on Points that are Learnable, Worth Learning, and Not Yet Learnt

Training on web-scale data can take months. But most computation and tim...

12 Sören Mindermann, et al. ∙

research

∙ 06/05/2022

Learning Dynamics and Generalization in Reinforcement Learning

Solving a reinforcement learning (RL) problem poses two competing challe...

0 Clare Lyle, et al. ∙

research

∙ 05/27/2022

Tranception: protein fitness prediction with autoregressive transformers and inference-time retrieval

The ability to accurately model the fitness landscape of protein sequenc...

0 Pascal Notin, et al. ∙

research

∙ 05/18/2022

Marginal and Joint Cross-Entropies Predictives for Online Bayesian Inference, Active Learning, and Active Sampling

Principled Bayesian deep learning (BDL) does not live up to its potentia...

36 Andreas Kirsch, et al. ∙

research

∙ 04/21/2022

Scalable Sensitivity and Uncertainty Analysis for Causal-Effect Estimates of Continuous-Valued Interventions

Estimating the effects of continuous-valued interventions from observati...

0 Andrew Jesson, et al. ∙

research

∙ 03/03/2022

Interventions, Where and How? Experimental Design for Causal Models at Scale

Causal discovery from observational and interventional data is challengi...

11 Panagiotis Tigas, et al. ∙

research

∙ 02/16/2022

Prospect Pruning: Finding Trainable Weights at Initialization using Meta-Gradients

Pruning neural networks at initialization would enable us to find sparse...

0 Milad Alizadeh, et al. ∙

research

∙ 02/14/2022

Active Surrogate Estimators: An Active Learning Approach to Label-Efficient Model Evaluation

We propose Active Surrogate Estimators (ASEs), a new method for label-ef...

3 Jannik Kossen, et al. ∙

research

∙ 02/03/2022

A Note on "Assessing Generalization of SGD via Disagreement"

Jiang et al. (2021) give empirical evidence that the average test error ...

2 Andreas Kirsch, et al. ∙

research

∙ 12/24/2021

DARTS without a Validation Set: Optimizing the Marginal Likelihood

The success of neural architecture search (NAS) has historically been li...

0 Miroslav Fil, et al. ∙

research

∙ 12/01/2021

Decomposing Representations for Deterministic Uncertainty Estimation

Uncertainty estimation is a key component in any deployed machine learni...

0 Haiwen Huang, et al. ∙

research

∙ 11/29/2021

DeDUCE: Generating Counterfactual Explanations Efficiently

When an image classifier outputs a wrong class label, it can be helpful ...

0 Benedikt Höltgen, et al. ∙

research

∙ 11/15/2021

Contrastive Representation Learning with Trainable Augmentation Channel

In contrastive representation learning, data representation is trained s...

0 Masanori Koyama, et al. ∙

research

∙ 11/05/2021

Multi-Spectral Multi-Image Super-Resolution of Sentinel-2 with Radiometric Consistency Losses and Its Effect on Building Delineation

High resolution remote sensing imagery is used in broad range of tasks, ...

0 Muhammed Razzak, et al. ∙

research

∙ 11/03/2021

Causal-BALD: Deep Bayesian Active Learning of Outcomes to Infer Treatment-Effects from Observational Data

Estimating personalized treatment effects from high-dimensional observat...

13 Andrew Jesson, et al. ∙

research

∙ 10/29/2021

Deep Deterministic Uncertainty for Semantic Segmentation

We extend Deep Deterministic Uncertainty (DDU), a method for uncertainty...

0 Jishnu Mukhoti, et al. ∙

research

∙ 10/28/2021

Using Non-Linear Causal Models to Study Aerosol-Cloud Interactions in the Southeast Pacific

Aerosol-cloud interactions include a myriad of effects that all begin wh...

0 Andrew Jesson, et al. ∙

research

∙ 10/22/2021

GeneDisco: A Benchmark for Experimental Design in Drug Discovery

In vitro cellular experimentation with genetic interventions, using for ...

19 Arash Mehrjou, et al. ∙

research

∙ 07/29/2021

Quantifying Uncertainty for Machine Learning Based Diagnostic

Virtual Diagnostic (VD) is a deep learning tool that can be used to pred...

0 Owen Convery, et al. ∙

research

∙ 07/15/2021

Shifts: A Dataset of Real Distributional Shift Across Multiple Large-Scale Tasks

There has been significant research done on developing methods for impro...

0 Andrey Malinin, et al. ∙

research

∙ 07/06/2021

Prioritized training on points that are learnable, worth learning, and not yet learned

We introduce Goldilocks Selection, a technique for faster model training...

5 Sören Mindermann, et al. ∙

research

∙ 06/30/2021

Improving black-box optimization in VAE latent space using decoder uncertainty

Optimization in the latent space of variational autoencoders is a promis...

0 Pascal Notin, et al. ∙

research

∙ 06/22/2021

A Practical Unified Notation for Information-Theoretic Quantities in ML

Information theory is of importance to machine learning, but the notatio...

51 Andreas Kirsch, et al. ∙

research

∙ 06/22/2021

A Simple Baseline for Batch Active Learning with Stochastic Acquisition Functions

In active learning, new labels are commonly acquired in batches. However...

5 Andreas Kirsch, et al. ∙

research

∙ 06/22/2021

Active Learning under Pool Set Distribution Shift and Noisy Data

Active Learning is essential for more label-efficient deep learning. Bay...

39 Andreas Kirsch, et al. ∙

research

∙ 06/14/2021

KL Guided Domain Adaptation

Domain adaptation is an important problem and often needed for real-worl...

0 A. Tuan Nguyen, et al. ∙

research

∙ 06/07/2021

Uncertainty Baselines: Benchmarks for Uncertainty Robustness in Deep Learning

High-quality estimates of uncertainty and robustness are crucial for num...

0 Zachary Nado, et al. ∙

research

∙ 06/04/2021

Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning

We challenge a common assumption underlying most supervised deep learnin...

19 Jannik Kossen, et al. ∙

research

∙ 06/04/2021

Can convolutional ResNets approximately preserve input distances? A frequency analysis perspective

ResNets constrained to be bi-Lipschitz, that is, approximately distance ...

0 Lewis Smith, et al. ∙

research

∙ 04/20/2021

Outcome-Driven Reinforcement Learning via Variational Inference

While reinforcement learning algorithms provide automated acquisition of...

0 Tim G. J. Rudner, et al. ∙

research

∙ 04/10/2021

Physically-Consistent Generative Adversarial Networks for Coastal Flood Visualization

As climate change increases the intensity of natural disasters, society ...

0 Björn Lütjens, et al. ∙

research

∙ 03/16/2021

Generating Interpretable Counterfactual Explanations By Implicit Minimisation of Epistemic and Aleatoric Uncertainties

Counterfactual explanations (CEs) are a practical tool for demonstrating...

10 Lisa Schut, et al. ∙

research

∙ 03/10/2021

Robustness to Pruning Predicts Generalization in Deep Neural Networks

Existing generalization measures that aim to capture a model's simplicit...

0 Lorenz Kuhn, et al. ∙

Yarin Gal

Featured Co-authors