English
Related papers

Related papers: Progress measures for grokking via mechanistic int…

200 papers

Explaining the predictions of neural black-box models is an important problem, especially when such models are used in applications where user trust is crucial. Estimating the influence of training examples on a learned neural model's…

Machine Learning · Computer Science 2020-12-16 Carolin Lawrence , Timo Sztyler , Mathias Niepert

This paper proposes an approach for segmenting a task consisting of compliant motions into phases, learning a primitive for each segmented phase of the task, and reproducing the task by sequencing primitives online based on the learned…

Robotics · Computer Science 2018-09-05 Tesfamichael Marikos Hagos , Markku Suomalainen , Ville Kyrki

Interpretability is crucial for ensuring RL systems align with human values. However, it remains challenging to achieve in complex decision making domains. Existing methods frequently attempt interpretability at the level of fundamental…

Machine Learning · Computer Science 2025-06-03 Anna Soligo , Pietro Ferraro , David Boyle

The study of mechanistic interpretability aims to reverse-engineer a model to explain its behaviors. While recent studies have focused on the static mechanism of a certain behavior, the learning dynamics inside a model remain to be…

Machine Learning · Computer Science 2025-09-24 Yueyan Li , Wenhao Gao , Caixia Yuan , Xiaojie Wang

We present PROGRESSOR, a novel framework that learns a task-agnostic reward function from videos, enabling policy training through goal-conditioned reinforcement learning (RL) without manual supervision. Underlying this reward is an…

Robotics · Computer Science 2024-11-28 Tewodros Ayalew , Xiao Zhang , Kevin Yuanbo Wu , Tianchong Jiang , Michael Maire , Matthew R. Walter

Universality is a key hypothesis in mechanistic interpretability -- that different models learn similar features and circuits when trained on similar tasks. In this work, we study the universality hypothesis by examining how small neural…

Machine Learning · Computer Science 2023-05-26 Bilal Chughtai , Lawrence Chan , Neel Nanda

Neural collapse, i.e., the emergence of highly symmetric, class-wise clustered representations, is frequently observed in deep networks and is often assumed to reflect or enable generalization. In parallel, flatness of the loss landscape…

Machine Learning · Computer Science 2026-02-05 Ting Han , Linara Adilova , Henning Petzka , Jens Kleesiek , Michael Kamp

Understanding neural networks is challenging due to their high-dimensional, interacting components. Inspired by human cognition, which processes complex sensory data by chunking it into recurring entities, we propose leveraging this…

Machine Learning · Computer Science 2025-02-05 Shuchen Wu , Stephan Alaniz , Eric Schulz , Zeynep Akata

Mechanistic interpretability has transformed the analysis of transformer circuits by decomposing model behavior into competing algorithms, identifying phase transitions during training, and deriving closed-form predictions for when and why…

Machine Learning · Computer Science 2026-03-19 Alma Lago

Transformation Synchronization is the problem of recovering absolute transformations from a given set of pairwise relative motions. Despite its usefulness, the problem remains challenging due to the influences from noisy and outlier…

Computer Vision and Pattern Recognition · Computer Science 2021-11-02 Zi Jian Yew , Gim Hee Lee

We attribute grokking, the phenomenon where generalization is much delayed after memorization, to compression. To do so, we define linear mapping number (LMN) to measure network complexity, which is a generalized version of linear region…

Machine Learning · Computer Science 2023-10-10 Ziming Liu , Ziqian Zhong , Max Tegmark

Neural networks trained with standard objectives exhibit behaviors characteristic of probabilistic inference: soft clustering, prototype specialization, and Bayesian uncertainty tracking. These phenomena appear across architectures -- in…

Machine Learning · Computer Science 2026-01-01 Alan Oursland

Modular exponentiation is crucial to number theory and cryptography, yet remains largely unexplored from a mechanistic interpretability standpoint. We train a 4-layer encoder-decoder Transformer model to perform this operation and…

Machine Learning · Computer Science 2025-10-24 David Demitri Africa , Sara M. Kapoor , Theo Simon Sorg , Challenger Mishra

This paper presents a novel neural network training approach for faster convergence and better generalization abilities in deep reinforcement learning. Particularly, we focus on the enhancement of training and evaluation performance in…

Machine Learning · Computer Science 2020-05-26 Mohammed Sharafath Abdul Hameed , Gavneet Singh Chadha , Andreas Schwung , Steven X. Ding

The capacity of foundation models allows for their application to new, unseen tasks. The adaptation to such tasks is called transfer learning. An efficient transfer learning method that circumvents parameter optimization is imprinting. The…

Machine Learning · Computer Science 2026-03-04 Justus Westerhoff , Golzar Atefi , Mario Koddenbrock , Alexei Figueroa , Alexander Löser , Erik Rodner , Felix A. Gers

Interpreting the learning dynamics of neural networks can provide useful insights into how networks learn and the development of better training and design approaches. We present an approach to interpret learning in neural networks by…

Machine Learning · Computer Science 2022-03-29 Ayush Manish Agrawal , Atharva Tendle , Harshvardhan Sikka , Sahib Singh

We study the dynamics of gradient flow with small weight decay on general training losses $F: \mathbb{R}^d \to \mathbb{R}$. Under mild regularity assumptions and assuming convergence of the unregularised gradient flow, we show that the…

Machine Learning · Computer Science 2025-11-06 Etienne Boursier , Scott Pesme , Radu-Alexandru Dragomir

Generative models have recently been explored for synthesizing neural network weights. These approaches take neural network checkpoints as training data and aim to generate high-performing weights during inference. In this work, we examine…

Machine Learning · Computer Science 2025-10-06 Boya Zeng , Yida Yin , Zhiqiu Xu , Zhuang Liu

Standard optimization theories struggle to explain grokking, where generalization occurs long after training convergence. While geometric studies attribute this to slow drift, they often overlook the interaction between the optimizer's…

Machine Learning · Computer Science 2026-03-17 Pratyush Acharya , Habish Dhakal

Mechanistic interpretability work attempts to reverse engineer the learned algorithms present inside neural networks. One focus of this work has been to discover 'circuits' -- subgraphs of the full model that explain behaviour on specific…

Machine Learning · Computer Science 2024-07-12 Joseph Miller , Bilal Chughtai , William Saunders
‹ Prev 1 4 5 6 7 8 10 Next ›