Unsupervised Learning: Foundations of Neural Computation

By Deliang Wang. 2001

318 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Representation & Self-Supervised Learning, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cognitive Neuroscience

In Representation & Self-Supervised Learning

In Signal Processing

Citations in the corpus

Data-Efficient Image Recognition with Contrastive Predictive Coding (2019, background): “structure of the world in an unsupervised manner (Barlow, 1989; Hinton et al., 1999; LeCun et al., 2015).”
Deep learning for smart fish farming: applications, opportunities and challenges (2020, background): “In unsupervised learning, the data are not labeled; instead the model seeks previously undetected patterns in a dataset with no pre-existing labels and with minimal human supervision (Geoffrey E Hinton, 1999).”
Machine Learning for 5G/B5G Mobile and Wireless Communications: Potential, Limitations, and Future Directions (2019, background): “The most common approach is to report the average logprobability that the model assigns to some example [182].”
Detecting Cyber Attacks in Industrial Control Systems Using Convolutional Neural Networks (2018, background): “Therefore, most recent research on ICS intrusion detection uses unsupervised (learning from unlabeled real data) [16] and semi-supervised (training using a set of partially labeled data, e.”
Let a biogeography-based optimizer train your Multi-Layer Perceptron (2014, background): “It may be divided into two types: supervised [53,74] and unsupervised [50,58] learning.”
Learn from Others and Be Yourself in Heterogeneous Federated Learning (2022, methodology): “For the heterogeneity problem, we take inspiration from the self-supervised learning [5, 6, 11, 13, 18, 25, 49, 91, 94].”
Multi-agent Reinforcement Learning: An Overview (2010, background): “The RL feedback is, however, more informative than in unsupervised learning, where there is no explicit feedback on the performance [104].”
The time dimension for scene analysis (2005, background): “A direct attempt to address figure-ground separation was made by Sejnowski and Hinton [113] based on a related learning algorithm—Boltzmann machine .”
Membership Inference Attacks on Machine Learning: A Survey (2021, methodology): “An unsupervised ML model aims to extract features and patterns from unlabeled data or labeled data without access to the labels [68].”
A survey of intrusion detection techniques for cyber-physical systems (2014, unknown): “Examples of collection are logging ACM Computing Surveys, Vol. 46, No. 4, Article 55, Publication date: March 2014.”
A Comprehensive Survey of Multiagent Reinforcement Learning (2008, unknown): “The RL feedback is, however, more informative than in unsupervised learning, where the agent would be left to discover the correct actions on its own, without any explicit feedback on its performance [10].”
Neuromodulated Spike-Timing-Dependent Plasticity, and Theory of Three-Factor Learning Rules (2016, methodology): “Unsupervised learning methods (e.g., Kohonen, 1990; Hinton and Sejnowski, 1999; Franzius et al., 2007) are one way to achieve compact representations of inputs in complex learning tasks (Arleo and Gerstner, 2000; Arleo et al., 2004; Swinehart and Abbott, 2006; Franzius et al., 2007; Sheynikhovich…”
To Cognize is to Categorize: Cognition is Categorization (2005, background): “This internal competition tends to bring into focus the structure inherent in and afforded by the input (Hinton & Sejnowsky 1999).”
Deep learning from temporal coherence in video (2009, background): “From a biological point of view, several authors (Hinton & Sejnowski, 1999; Becker, 1996a) argue that pure supervised learning is a poor model of how animals really learn.”
MixCo: Mix-up Contrastive Learning for Visual Representation (2020, background): “Learning representations without human annotations on data is very promising since it enables us to utilize a massive amount of data without demanding effort to label them [15].”