English
Related papers

Related papers: Small-scale proxies for large-scale Transformer tr…

200 papers

Underpinning the past decades of work on the design, initialization, and optimization of neural networks is a seemingly innocuous assumption: that the network is trained on a \textit{stationary} data distribution. In settings where this…

Machine Learning · Computer Science 2024-03-01 Clare Lyle , Zeyu Zheng , Khimya Khetarpal , Hado van Hasselt , Razvan Pascanu , James Martens , Will Dabney

Most existing analyses of (stochastic) gradient descent rely on the condition that for $L$-smooth costs, the step size is less than $2/L$. However, many works have observed that in machine learning applications step sizes often do not…

Optimization and Control · Mathematics 2022-06-10 Kwangjun Ahn , Jingzhao Zhang , Suvrit Sra

Scaling distributed GPU training is commonly assumed to yield predictable performance gains as additional nodes are added. In practice, many large-scale deployments encounter diminishing returns and unstable behavior well before theoretical…

Networking and Internet Architecture · Computer Science 2026-03-06 Dinesh Gopalan , Ratul Ali

Fine-tuning a pre-trained model (such as BERT, ALBERT, RoBERTa, T5, GPT, etc.) has proven to be one of the most promising paradigms in recent NLP research. However, numerous recent works indicate that fine-tuning suffers from the…

Machine Learning · Computer Science 2023-12-08 Zihao Fu , Anthony Man-Cho So , Nigel Collier

Transformers trained in low precision can suffer forward-error amplification. We give a first-order, module-wise theory that predicts when and where errors grow. For self-attention we derive a per-layer bound that factorizes into three…

Machine Learning · Computer Science 2025-10-28 Jinwoo Baek

Modulation instability is a phenomenon of spontaneous pattern formation in nonlinear media, oftentimes leading to an unpredictable behaviour and a degradation of a signal of interest. We propose an approach based on reinforcement learning…

Pattern Formation and Solitons · Physics 2024-07-24 Nikolay Kalmykov , Rishat Zagidullin , Oleg Rogov , Sergey Rykovanov , Dmitry V. Dylov

This paper considers optimization problems over networks where agents have individual objectives to meet, or individual parameter vectors to estimate, subject to subspace constraints that require the objectives across the network to lie in…

Multiagent Systems · Computer Science 2020-04-22 Roula Nassif , Stefan Vlaski , Ali H. Sayed

Downstream scaling laws aim to predict task performance at larger scales from the model's performance at smaller scales. Whether such prediction should be possible is unclear: some works discover clear linear scaling trends after simple…

Computation and Language · Computer Science 2025-10-10 Nicholas Lourie , Michael Y. Hu , Kyunghyun Cho

Large Language Models (LLMs) are pretrained on massive datasets and later instruction-tuned via supervised fine-tuning (SFT) or reinforcement learning (RL). Best practices emphasize large, diverse pretraining data, whereas post-training…

Machine Learning · Computer Science 2026-03-03 Adel Javanmard , Baharan Mirzasoleiman , Vahab Mirrokni

Transformers have recently revolutionized many domains in modern machine learning and one salient discovery is their remarkable in-context learning capability, where models can solve an unseen task by utilizing task-specific prompts without…

Machine Learning · Computer Science 2023-10-10 Yu Huang , Yuan Cheng , Yingbin Liang

In this paper, we find a new large scale instability displayed by a stratified rotating flow in forced turbu- lence. The turbulence is generated by a small scale external force at low Reynolds number. The theory is built on the rigorous…

Fluid Dynamics · Physics 2013-11-13 Anatoly Tur , Malik Chabane , Vladimir Yanovsky

Traditional analyses of gradient descent optimization show that, when the largest eigenvalue of the loss Hessian - often referred to as the sharpness - is below a critical learning-rate threshold, then training is 'stable' and training loss…

Machine Learning · Computer Science 2024-12-24 Lawrence Wang , Stephen J. Roberts

We identify stable regions in the residual stream of Transformers, where the model's output remains insensitive to small activation changes, but exhibits high sensitivity at region boundaries. These regions emerge during training and become…

Machine Learning · Computer Science 2024-11-19 Jett Janiak , Jacek Karwowski , Chatrik Singh Mangat , Giorgi Giglemiani , Nora Petrova , Stefan Heimersheim

Transformer models have emerged as fundamental tools across various scientific and engineering disciplines, owing to their outstanding performance in diverse applications. Despite this empirical success, the theoretical foundations of…

Machine Learning · Computer Science 2026-04-14 Zhen Qin , Jinxin Zhou , Jiachen Jiang , Zhihui Zhu

While scaling laws for Large Language Models (LLMs) traditionally focus on proxy metrics like pretraining loss, predicting downstream task performance has been considered unreliable. This paper challenges that view by proposing a direct…

Machine Learning · Computer Science 2025-12-10 Jakub Krajewski , Amitis Shidani , Dan Busbridge , Sam Wiseman , Jason Ramapuram

There has been a recent interest in imitation learning methods that are guaranteed to produce a stabilizing control law with respect to a known system. Work in this area has generally considered linear systems and controllers, for which…

Optimization and Control · Mathematics 2021-09-23 Sebastian East

Time-dependent data-generating distributions have proven to be difficult for gradient-based training of neural networks, as the greedy updates result in catastrophic forgetting of previously learned knowledge. Despite the progress in the…

Machine Learning · Computer Science 2023-04-03 Matthias De Lange , Gido van de Ven , Tinne Tuytelaars

Recent developments in large-scale machine learning suggest that by scaling up data, model size and training time properly, one might observe that improvements in pre-training would transfer favorably to most downstream tasks. In this work,…

Machine Learning · Computer Science 2021-10-06 Samira Abnar , Mostafa Dehghani , Behnam Neyshabur , Hanie Sedghi

Within the scaling laws paradigm, which underpins the training of large neural networks like ChatGPT and Llama, we consider a supervised regression setting and establish the existance of a strong form of the model collapse phenomenon, a…

Machine Learning · Computer Science 2024-10-10 Elvis Dohmatob , Yunzhen Feng , Arjun Subramonian , Julia Kempe

Imitation learning has achieved great success in many sequential decision-making tasks, in which a neural agent is learned by imitating collected human demonstrations. However, existing algorithms typically require a large number of…

Machine Learning · Computer Science 2023-06-14 Tianxiang Zhao , Wenchao Yu , Suhang Wang , Lu Wang , Xiang Zhang , Yuncong Chen , Yanchi Liu , Wei Cheng , Haifeng Chen