Canlyniadau Chwilio - Kirsch, Andreas

Speeding Up BatchBALD: A k-BALD Family of Approximations for Active Learning gan Kirsch, Andreas

Cael y testun llawn
Cael y testun llawn

Black-Box Batch Active Learning for Regression gan Kirsch, Andreas

Cael y testun llawn
Cael y testun llawn

Does "Deep Learning on a Data Diet" reproduce? Overall yes, but GraNd at Initialization does not gan Kirsch, Andreas

Cael y testun llawn
Cael y testun llawn

Advancing Deep Active Learning & Data Subset Selection: Unifying Principles with Information-Theory Intuitions gan Kirsch, Andreas

Cael y testun llawn
Cael y testun llawn

(Implicit) Ensembles of Ensembles: Epistemic Uncertainty Collapse in Large Models gan Kirsch, Andreas

Cael y testun llawn
Cael y testun llawn

LSTM-Style Looped Transformers with Shared KV Caches gan Kirsch, Andreas

Cael y testun llawn
Cael y testun llawn

PowerEvaluationBALD: Efficient Evaluation-Oriented Deep (Bayesian) Active Learning with Stochastic Acquisition Functions gan Kirsch, Andreas, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

A Practical & Unified Notation for Information-Theoretic Quantities in ML gan Kirsch, Andreas, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

A Note on "Assessing Generalization of SGD via Disagreement" gan Kirsch, Andreas, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

Unifying Approaches in Active Learning and Active Sampling via Fisher Information and Information-Theoretic Quantities gan Kirsch, Andreas, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

The PML-Method for a Scattering Problem for a Local Perturbation of an Open Periodic Waveguide gan Kirsch, Andreas, Zhang, Ruming

Cael y testun llawn
Cael y testun llawn

Direct and inverse time-harmonic scattering by Dirichlet periodic curves with local perturbations gan Hu, Guanghui, Kirsch, Andreas

Cael y testun llawn
Cael y testun llawn

Unpacking Information Bottlenecks: Unifying Information-Theoretic Objectives in Deep Learning gan Kirsch, Andreas, Lyle, Clare, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

Test Distribution-Aware Active Learning: A Principled Approach Against Distribution Shift and Outliers gan Kirsch, Andreas, Rainforth, Tom, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

Marginal and Joint Cross-Entropies & Predictives for Online Bayesian Inference, Active Learning, and Active Sampling gan Kirsch, Andreas, Kossen, Jannik, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

The Benefits and Risks of Transductive Approaches for AI Fairness gan Razzak, Muhammed, Kirsch, Andreas, Gal, Yarin

Cael y testun llawn
Cael y testun llawn

When three experiments are better than two: Avoiding intractable correlated aleatoric uncertainty by leveraging a novel bias--variance tradeoff gan Scherer, Paul, Kirsch, Andreas, Taylor-King, Jake P.

Cael y testun llawn
Cael y testun llawn

CoLoR-Filter: Conditional Loss Reduction Filtering for Targeted Language Model Pre-training gan Brandfonbrener, David, Zhang, Hanlin, Kirsch, Andreas, Schwarz, Jonathan Richard, Kakade, Sham

Cael y testun llawn
Cael y testun llawn

Large Language Model Selection with Limited Annotations gan Durmazkeser, Yavuz, Okanovic, Patrik, Kirsch, Andreas, Hoefler, Torsten, Gürel, Nezihe Merve

Cael y testun llawn
Cael y testun llawn

Large Language Model Selection with Limited Annotations gan Durmazkeser, Yavuz, Okanovic, Patrik, Kirsch, Andreas, Hoefler, Torsten, Gürel, Nezihe Merve

Cael y testun llawn
Cael y testun llawn