paper-with-me

Papers

Transfer Learning Beyond the Standard Model

2025-10-22 · Veena Krishnaraj, Adrian E. Bayer, Christian Kragh Jespersen, Peter Melchior arxiv

Machine learning enables powerful cosmological inference but typically requires many high-fidelity simulations covering many cosmological models. Transfer learning offers a way to reduce the simulation cost by reusing knowledge across models. We show that pre-training on the standard model of cosmology, $Λ$CDM, and fine-tuning on various beyond-$Λ$CDM scenarios -- including massive neutrinos, modified gravity, and primordial non-Gaussianities -- can enable inference with significantly fewer beyond-$Λ$CDM simulations. However, we also show that negative transfer can occur when strong physical degeneracies exist between $Λ$CDM and beyond-$Λ$CDM parameters. We consider various transfer architectures, finding that including bottleneck structures provides the best performance. Our findings illustrate the opportunities and pitfalls of foundation-model approaches in physics: pre-training can accelerate inference, but may also hinder learning new physics.

📄 PDF Abstract BibTeX arXiv:2510.19168

Code (0)

등록된 구현이 없습니다.

Tasks

Transfer Learning

Similar Papers 제목 키워드 기반

On Transfer in Classification: How Well do Subsets of Classes Generalize?

2024-03-06 · Raphael Baena, Lucas Drumetz, Vincent Gripon

In classification, it is usual to observe that models trained on a given set of classes can generalize to previously unseen ones, suggesting the ability to learn beyond the initial task. This ability is often leveraged i…

Few-Shot LearningTransfer Learning

Beyond Perception Errors: Semantic Fixation in Large Vision-Language Models

2026-04-13 · Md Tanvirul Alam arxiv

Large vision-language models (VLMs) often rely on familiar semantic priors, but existing evaluations do not cleanly separate perception failures from rule-mapping failures. We study this behavior as semantic fixation: pr…

Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

2022-06-09 · Tianlong Chen, Zhenyu Zhang, Sijia Liu, Yang Zhang 외

Pre-training serves as a broadly adopted starting point for transfer learning on various downstream tasks. Recent investigations of lottery tickets hypothesis (LTH) demonstrate such enormous pre-trained models can be rep…

Transfer Learning

Testable Likelihoods for Beyond-the-Standard Model Fits

2023-09-19 · Anja Beck, Méril Reboud, Danny van Dyk

Studying potential BSM effects at the precision frontier requires accurate transfer of information from low-energy measurements to high-energy BSM models. We propose to use normalising flows to construct likelihood funct…

FormmodelNormalising Flows

Beyond Fine-Tuning: Transferring Behavior in Reinforcement Learning

2021-02-24 · Víctor Campos, Pablo Sprechmann, Steven Hansen, Andre Barreto 외

Designing agents that acquire knowledge autonomously and use it to solve new tasks efficiently is an important challenge in reinforcement learning. Knowledge acquired during an unsupervised pre-training phase is often tr…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)Unsupervised Pre-training