This article highlights specific features of biological neurons and their dendritic trees, whose adoption may help advance artificial neural networks used in various machine learning applications. Advancements could take the form of increased computational capabilities and/or reduced power consumption. Proposed features include dendritic anatomy, dendritic nonlinearities, and compartmentalized plasticity rules, all of which shape learning and information processing in biological networks. We discuss the com
From handcrafted feature detectors to vision transformers, we've come a long way...
What is a Convolutional Neural Network? A Convolutional Neural Network (CNN) is a specialized type of artificial neural network that is primarily designed for processing and analyzing structured grid-like data, such as images and videos. CNNs have revolutionized the field of computer vision and are widely used for tasks such as image classification, object detection
Graph neural networks (GNNs) have proven their efficacy in a variety of real-world applications, but their underlying mechanisms remain a mystery. To address this challenge and enable reliable decision-making, many GNN explainers have been proposed in recent years. However, these methods often encounter limitations, including their dependence on specific instances, lack of generalizability to unseen graphs, producing potentially invalid explanations, and yielding inadequate fidelity. To overcome these limit
Abstract page for arXiv paper 2501.07451v1: A Survey on Dynamic Neural Networks: from Computer Vision to Multi-modal Sensor Fusion
This explores whether the well-known finding that neural networks store concepts as straight-line directions is a real property of the networks — or an illusion created by the fact that the tools we u
OOPS primitives for recurrent neural networks (Matteo Gagliolo, IDSIA, ongoing) RNN programming language: not based on stacks and traditional languages, but on matrix multiplications, simple local weight change algorithms, primitives for network growth etc.. Back to J. Schmidhuber 's OOPS page
NIPS 2017 Mon Dec 4th through Sat the 9th, 2017 at Long Beach Convention Center Paper ID: 617 Title: Self-Normalizing Neural Networks Reviewer 1 I am putting "accept" because this paper already seems to be attracting a lot of attention in the DL community and I see no reason to squash it. However I do have some reservations. There is all this theory and a huge derivation, showing that under certain conditions this particular type of unit will lead to activation norms that don't blow up or get small. Basical
# Bayesian Networks (6) - DeepMind Relies on this Old Statistical Method to Build Fair Machine Learning Models - Oct 23, 2020. Causal Bayesian Networks are used to model the influence of fairness attributes in a dataset. DeepMind is Using This Old Technique to Evaluate Fairness in Machine Learning Models - Oct 28, 2019. Visualizing the datasets is an essential component to identify potential sources of bias and unfairness. DeepMind relied on a method called Causal Bayesian networks (CBNs) to represent a
Inducing Causal Structure for Interpretable Neural NetworksAtticus Geiger, Zhengxuan Wu, Hanson Lu, Josh Rozner, Elisa Kreiss, Thoma