English
Related papers

Related papers: Localizing Model Behavior with Path Patching

200 papers

Landmark-based human action recognition in videos is a challenging task in computer vision. One key step is to design a generic approach that generates discriminative features for the spatial structure and temporal dynamics. To this end, we…

Computer Vision and Pattern Recognition · Computer Science 2019-12-13 Weixin Yang , Terry Lyons , Hao Ni , Cordelia Schmid , Lianwen Jin

Traditional deep network training methods optimize a monolithic objective function jointly for all the components. This can lead to various inefficiencies in terms of potential parallelization. Local learning is an approach to…

Machine Learning · Computer Science 2023-01-19 Adeetya Patel , Michael Eickenberg , Eugene Belilovsky

Path-following algorithms are frequently used in composite optimization problems where a series of subproblems, with varying regularization hyperparameters, are solved sequentially. By reusing the previous solutions as initialization,…

Optimization and Control · Mathematics 2021-12-10 Eugene Ndiaye , Ichiro Takeuchi

Unveil the homophilic/heterophilic behaviors that characterize the wiring patterns of complex networks is an important task in social network analysis, often approached studying the assortative mixing of node attributes. Recent works…

Social and Information Networks · Computer Science 2020-12-10 Giulio Rossetti , Salvatore Citraro , Letizia Milli

Graph Neural Networks (GNNs), despite achieving remarkable performance across different tasks, are theoretically bounded by the 1-Weisfeiler-Lehman test, resulting in limitations in terms of graph expressivity. Even though prior works on…

Machine Learning · Computer Science 2024-04-02 Quang Truong , Peter Chin

Networks play a central role in modern data analysis, enabling us to reason about systems by studying the relationships between their parts. Most often in network analysis, the edges are given. However, in many systems it is difficult or…

Machine Learning · Statistics 2014-02-06 Scott W. Linderman , Ryan P. Adams

Passive network tomography uses end-to-end observations of network communication to characterize the network, for instance to estimate the network topology and to localize random or adversarial glitches. Under the setting of linear network…

Networking and Internet Architecture · Computer Science 2016-11-15 Hongyi Yao , Sidharth Jaggi , Minghua Chen

Process mining leverages event data extracted from IT systems to generate insights into the business processes of organizations. Such insights benefit from explicitly considering the frequency of behavior in business processes, which is…

Formal Languages and Automata Theory · Computer Science 2025-07-10 Tian Li , Artem Polyvyanyy , Sander J. J. Leemans

We present a new framework for Patch Distribution Modeling, PaDiM, to concurrently detect and localize anomalies in images in a one-class learning setting. PaDiM makes use of a pretrained convolutional neural network (CNN) for patch…

Computer Vision and Pattern Recognition · Computer Science 2020-11-18 Thomas Defard , Aleksandr Setkov , Angelique Loesch , Romaric Audigier

This work focuses on the generic identifiability of dynamical networks with partial excitation and measurement: a set of nodes are interconnected by transfer functions according to a known topology, some nodes are excited, some are…

Optimization and Control · Mathematics 2021-10-04 Antoine Legat , Julien M. Hendrickx

Community detection, a fundamental task for network analysis, aims to partition a network into multiple sub-structures to help reveal their latent functions. Community detection has been extensively studied in and broadly applied to many…

Social and Information Networks · Computer Science 2021-08-17 Di Jin , Zhizhi Yu , Pengfei Jiao , Shirui Pan , Dongxiao He , Jia Wu , Philip S. Yu , Weixiong Zhang

Characterizing the community structure of complex networks is a key challenge in many scientific fields. Very diverse algorithms and methods have been proposed to this end, many working reasonably well in specific situations. However, no…

Physics and Society · Physics 2013-01-01 Rodrigo Aldecoa , Ignacio Marín

Dynamic networks consist of a sequence of time-varying networks, and it is of great importance to detect the network change points. Most existing methods focus on detecting abrupt change points, necessitating the assumption that the…

Methodology · Statistics 2023-10-13 Yuzhao Zhang , Jingnan Zhang , Yifan Sun , Junhui Wang

To operate in real-world high-stakes environments, deep learning systems have to endure noises that have been continuously thwarting their robustness. Data-end defense, which improves robustness by operations on input data instead of…

Computer Vision and Pattern Recognition · Computer Science 2022-04-14 Jiakai Wang , Zixin Yin , Pengfei Hu , Aishan Liu , Renshuai Tao , Haotong Qin , Xianglong Liu , Dacheng Tao

There is a wide range of topologies to use in simulation that can make research divergency; therefore, we propose a topology set that can be used in research of network behaviour in Software Defined Network (SDN). This paper can unite the…

Networking and Internet Architecture · Computer Science 2021-07-08 Alireza Shirmarz

Link prediction (LP) is an important problem in network science and machine learning research. The state-of-the-art LP methods are usually evaluated in a uniform setup, ignoring several factors associated with the data and application…

Social and Information Networks · Computer Science 2025-07-21 Bhargavi Kalyani , A Rama Prasad Mathi , Niladri Sett

Information of localization properties of eigenvectors of the complex network has applicability in many different areas which include networks centrality measures, spectral partitioning, development of approximation algorithms, and disease…

Physics and Society · Physics 2021-09-29 Priodyuti Pradhan , Sarika Jalan

This thesis surveys the research in patch-based synthesis and algorithms for finding correspondences between small local regions of images. We additionally explore a large kind of applications of this new fast randomized matching technique.…

Graphics · Computer Science 2020-05-14 Hadi Abdi Khojasteh

Among the several topological properties of complex networks, the shortest path represents a particularly important characteristic because of its potential impact not only on other topological properties, but mainly for its influence on…

Social and Information Networks · Computer Science 2020-03-30 Guilherme S. Domingues , Cesar H. Comin , Luciano da F. Costa

I start by reviewing some basic properties of random graphs. I then consider the role of random walks in complex networks and show how they may be used to explain why so many long tailed distributions are found in real data sets. The key…

Statistical Mechanics · Physics 2012-12-11 T. S. Evans