Abstract page for arXiv paper 1910.07969: On Completeness-aware Concept-Based Explanations in Deep Neural Networks
Rigorous treatment of GNNs: message passing framework, GCN spectral derivation, GAT attention, GraphSAGE sampling, WL expressivity limits, over-smoothing, and applications.
Society Leadership Founding Generation Fellowship Guidelines Merit Awards Award Recipients ICIS Listserv Congress 2026 Panama City Abstract Submissions Destination Pre-Congress Workshops Presentation Guidelines Program Registration Review Panels Speakers Sponsor/Exhibit Travel Awards Past Congresses Events & Initiatives Google Translate disclaimer Uncovering cognition in young infants using deep neural networks and awake fMRI Anyone who interacts with a young infant may find themselves wondering
Skip to content TheLinuxCode Software Menu Toggle Distros Menu Toggle SysAdmin Menu Toggle Residential Proxies Residential Proxies TheLinuxCode Main Menu Menu Optimization Rule in Deep Neural Networks: Practical Guide, 2026 Edition Leave a Comment / By Linux Code / January 8, 2026 Why I care about the optimization rule I build and ship deep learning systems weekly, and the single thing that decides whether a model trains in 2 hours or 2 weeks is the optimization rule. I treat the optimization rule like engi
NeurIPS Proceedings Search On Completeness-aware Concept-Based Explanations in Deep Neural Networks Chih-Kuan Yeh, Been Kim, Sercan Arik, Chun-Liang Li, Tomas Pfister, Pradeep K. Ravikumar Advances in Neural Information Processing Systems 33 (NeurIPS 2020) Abstract Human explanations of high-level decisions are often expressed in terms of key concepts the decisions are based on. In this paper, we study such concept-based explainability for Deep Neural Networks (DNNs). First, we define the notion of \emph{co
Posted by David Ha, Staff Research Scientist and Yujin Tang, Research Software Engineer, Google Research, Tokyo
| ABOUT | BACKGROUND || --> VISUALIZATION | --> ALGORITHM | CREDITS | Hidden Networks is a large-scale multi-channel video installation that applies machine learning techniques to the analysis of the moving image. A system of deep neural networks analyzes the optical flow in a film dataset and identifies scenes with similar motions. The dataset consists of two silent French film series, Les Vampires (1915) and Judex (1916), both directed by Louis Feuillade, a master of deep staging and visual choreography
The study of Neural Tangent Kernels (NTKs) has provided much needed insight into convergence and generalization properties of neural networks in the over-parametrized (wide) limit by approximating the network using a first-order Taylor expansion with respect to its weights in the neighborhood of their initialization values. This allows neural network training to be analyzed from the perspective of reproducing kernel Hilbert spaces (RKHS), which is informative in the over-parametrized regime, but a poor appr
Deep Learning application using Tensorflow and Keras Deep Convolutional Neural Networks changed the research landscape
The Gaussian activation function shapes the output of a neuron into a bell-shaped curve that is symmetric about its unique peak.