Score-of-Mixture Training: One-Step Generative Model Training Made Simple via Score Estimation of Mixture Distributions
A Unified View on Learning Unnormalized Distributions via Noise-Contrastive Estimation
Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search
Universal Features for High-Dimensional Learning and Inference
Authors: Shao-Lun Huang, Anuran Makur, Gregory W. Wornell and Lizhong Zheng
Publisher: Foundations and Trends® in Communications and Information Theory, Vol. 21, No. 1-2, pp.1-299. Now Publishers: Boston - Delft
Date Published: January 1, 2024
In many contemporary and emerging applications of machine learning and statistical inference, the phenomena of interest are characterized by variables defined over large alphabets. This increasing size of both the data and the number of inferences, and the limited available training data means there is a need to understand which inference tasks can be most effectively carried out, and, in turn, what features of the data are most relevant to them.
In this monograph, the authors develop the idea of extracting “universally good” features, and establish that diverse notions of such universality lead to precisely the same features. The information-theoretic approach used results in a local information geometric analysis that facilitates their computation in a host of applications.
The authors provide a comprehensive treatment that guides the reader through the basic principles to the advanced techniques including many new results. They emphasize a development from first-principles together with common, unifying terminology and notation, and pointers to the rich embodying literature, both historical and contemporary.
Written for students and researchers, this monograph is a complete treatise on the information theoretic treatment of a recognized and current problem in machine learning and statistical inference.
Systems and Methods of Detecting Moving Obstacles
Multi-User Communication System Employing Spread Signatures
Spread-Response Precoding System for Wireless Transmission
Lapped Orthogonal Vector Quantization
System, Method, and Product for Information Embedding Using an Ensemble of Non-Intersecting Embedding Generators
System, Method, and Product for Information Embedding Using an Ensemble of Non-Intersecting Embedding Generators
System, Method, and Product for Distortion-Compensated Information Embedding Using an Ensemble of Non-Intersecting Embedding Generators
Method and System for Packet Communication Employing Path Diversity
Block-iterative equalizers for digital communication system
Lossy data compression exploiting distortion side information
System and Apparatus for Error Control Codes Based on Layering and Linear Transformations
Systems and Method for Data Compression with Model-Free Encoding
Multi-Source Transfer Learning From Pre-Trained Networks
Group Fairness with Uncertainty in Sensitive Attributes
On Data-Driven Underwater Acoustic Direct Localization: Design Considerations of a Deep Neural Network-Based Solution
An Architecture for Passive Joint Localization and Environment Learning in Shallow-Water Underwater Acoustic Settings
On Counterfactual Inference With Unobserved Confounding
Few-Shot Transfer Learning from Multiple Pre-Trained Networks
A Real-Time Brain-Machine Interface for Enhancement of Sequential Motor Function
Signal Processing with Fractals: A Wavelet-Based Approach
Authors: G. W. Wornell
Publisher: Prentice-Hall: Upper Saddle River, NJ
Date Published: January 1, 1996
Chapters: 13
Fractal geometry and recent developments in wavelet theory are having an important impact on the field of signal processing. Efficient representations for fractal signals based on wavelets are opening up new applications for signal processing, and providing better solutions to problems in existing applications. Signal Processing with Fractals provides a valuable introduction to this new and exciting area, and develops a powerful conceptual foundation for understanding the topic. Practical techniques for synthesizing, analyzing, and processing fractal signal for a wide range of applications are developed in detail, and novel applications in communications are explored. Written by a signal processor for signal processors, Signal Processing with Fractals is a self-contained, well-illustrated treatise, and includes a highly accessible concept-oriented primer of the relevant wavelet theory.
Wireless Communications: Signal Processing Perspectives
Authors: H. V. Poor and G. W. Wornell, eds.
Publisher: Prentice-Hall: Upper Saddle River, NJ
Date Published: January 1, 1998
Chapters: 15
Signal processing algorithms and architectures have an increasingly important role to play in meeting the central challenges faced in the design of advanced wireless communication systems. In Wireless Communications: Signal Processing Perspectives, leaders in the field describe state-of-the-art research in applying signal processing methodologies in the context of tomorrow’s most important wireless applications, ranging from next-generation cellular telephony and personal communication services, to nomadic computing and wireless multimedia.
Wireless Communications: Signal Processing Perspectives is a valuable reference both for signal processing specialists seeking to apply their expertise in the rapidly growing wireless communications field, and for communications specialists eager to exploit signal processing techniques and implementations in developing efficient wireless systems of the future.
Wireless Communications: Signal Processing Perspectives includes both physical and network layer topics, and also contains a thought-provoking essay by Andrew J. Viterbi on the laws of nature and society that ultimately govern wireless networks.