IEEE Signal Processing Magazine最新文献_第2页

Low-Rank Tensor Decompositions for the Theory of Neural Networks: Understanding deep learning through the lens of tensor formats [Special Issue on the Mathematics of Deep Learning] 神经网络理论的低秩张量分解：从张量格式的角度理解深度学习[深度学习数学专刊]

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2025.3646570

Ricardo Borsoi;Konstantin Usevich;Marianne Clausel

{"title":"Low-Rank Tensor Decompositions for the Theory of Neural Networks: Understanding deep learning through the lens of tensor formats [Special Issue on the Mathematics of Deep Learning]","authors":"Ricardo Borsoi;Konstantin Usevich;Marianne Clausel","doi":"10.1109/MSP.2025.3646570","DOIUrl":"https://doi.org/10.1109/MSP.2025.3646570","url":null,"abstract":"The groundbreaking performance of deep neural networks (NNs) promoted a surge of interest in providing a mathematical basis to deep learning theory. Low-rank tensor decompositions are specially fit for this task due to their close connection to NNs and their rich theoretical results. Different tensor decompositions have strong uniqueness guarantees, which allow for a direct interpretation of their factors, and polynomial time algorithms have been proposed to compute them. Through the connections between tensors and NNs, such results supported many important advances in the theory of NNs. In this review article, we show how low-rank tensor methods—which have been a core tool in the signal processing and machine learning communities—play a fundamental role in theoretically explaining different aspects of the performance of deep NNs, including their expressivity, algorithmic learnability and computational hardness, generalization, and identifiability. Our goal is to give an accessible overview of existing approaches (developed by different communities, ranging from computer science to mathematics) in a coherent and unified way and to open a broader perspective on the use of low-rank tensor decompositions for the theory of deep NNs.","PeriodicalId":13246,"journal":{"name":"IEEE Signal Processing Magazine","volume":"43 2","pages":"107-121"},"PeriodicalIF":9.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147665361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Mathematical Foundations of Spiking Neural Networks: Strengths, challenges, and computational paradigm potential [Special Issue on the Mathematics of deep Learning] 脉冲神经网络的数学基础：优势、挑战和计算范式潜力[深度学习数学特刊]

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2025.3597033

Adalbert Fono;Manjot Singh;Ernesto Araya;Philipp C. Petersen;Holger Boche;Gitta Kutyniok

引用次数: 0

Guest Editorial for Part 1 of the Special Issue on the Mathematics of Deep Learning [From the Guest Editors] 深度学习数学特刊第1部分客座编辑[来自客座编辑]

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2026.3668468

Laura Balzano;Joan Bruna Estrach;Gitta Kutyniok;Robert Nowak;Jong Chul Ye

引用次数: 0

SPS Social Media SPS社交媒体

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2026.3674140

引用次数: 0

Understanding Untrained Deep Models for Inverse Problems [From the Guest Editors]: Algorithms and theory 理解反问题的未训练深度模型[来自客座编辑]：算法和理论

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2025.3632786

Ismail Alkhouri;Evan Bell;Avrajit Ghosh;Shijun Liang;Rongrong Wang;Saiprasad Ravishankar

{"title":"Understanding Untrained Deep Models for Inverse Problems [From the Guest Editors]: Algorithms and theory","authors":"Ismail Alkhouri;Evan Bell;Avrajit Ghosh;Shijun Liang;Rongrong Wang;Saiprasad Ravishankar","doi":"10.1109/MSP.2025.3632786","DOIUrl":"https://doi.org/10.1109/MSP.2025.3632786","url":null,"abstract":"In recent years, deep learning (DL) methods have been extensively developed for inverse imaging problems (IIPs), encompassing supervised, self-supervised, and generative approaches. Most of these methods require large numbers of labeled or unlabeled training data to learn effective models. However, in many practical applications, such as medical image reconstruction, extensive training datasets are often unavailable or limited. A significant milestone in addressing this challenge came in 2018 with the work of Ulyanov et al. <xref>[1]</xref>, which introduced the deep image prior (DIP)—the first training-data-free convolutional neural network (CNN) method for IIPs. Unlike conventional DL approaches, DIP requires only a CNN, the noisy measurements, and a forward operator. By leveraging the implicit regularization of deep networks initialized with random noise, DIP can learn and restore image structures without relying on external datasets. However, a well-known limitation of DIP is its susceptibility to overfitting, primarily due to overparameterization of the network. In this tutorial article, we provide a comprehensive review of DIP, including a theoretical analysis of its training dynamics. We also categorize and discuss recent advancements in DIP-based methods that are aimed at mitigating overfitting, including techniques such as regularization, network reparameterization, and early stopping. Furthermore, we discuss approaches that combine DIP with pretrained neural networks (NNs), present empirical comparison results against data-centric methods, and highlight open research questions and future directions.","PeriodicalId":13246,"journal":{"name":"IEEE Signal Processing Magazine","volume":"43 2","pages":"77-92"},"PeriodicalIF":9.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147665500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ILN - Transformer Architectures for Multimodal Signal Processing & Decision Making 用于多模态信号处理和决策的变压器架构

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2026.3673028

引用次数: 0

Advance Your Signal Processing Education 推进你的信号处理教育

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2026.3673053

引用次数: 0

In Memoriam: Edward R. Dougherty [In Memoriam] 纪念：爱德华·多尔蒂[纪念]

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-03-01 Epub Date: 2026-04-13 DOI: 10.1109/MSP.2026.3663797

Ulisses Braga-Neto;Yidong Chen

引用次数: 0

Artificial Intelligence Foundations: Building essential machine learning skills and understanding [Special Issue on Artificial Intelligence for Education: A Signal Processing Perspective] 人工智能基础：建立基本的机器学习技能和理解[教育人工智能特刊：信号处理视角]

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-02-18 DOI: 10.1109/MSP.2025.3604572

Kurt Butler;Mónica F. Bugallo;Petar M. Djurić

引用次数: 0

Future-Proofing Programmers: Optimal knowledge tracing for artificial intelligence-assisted personalized education [Special Issue on Artificial Intelligence for Education: A Signal Processing Perspective] 面向未来的程序员：人工智能辅助个性化教育的最佳知识追踪[人工智能教育特刊：信号处理视角]

IF 9.6 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-02-18 DOI: 10.1109/MSP.2025.3609896

Yuchen Wang;Pei-Duo Yu;Chee Wei Tan

{"title":"Future-Proofing Programmers: Optimal knowledge tracing for artificial intelligence-assisted personalized education [Special Issue on Artificial Intelligence for Education: A Signal Processing Perspective]","authors":"Yuchen Wang;Pei-Duo Yu;Chee Wei Tan","doi":"10.1109/MSP.2025.3609896","DOIUrl":"https://doi.org/10.1109/MSP.2025.3609896","url":null,"abstract":"Learning to learn is becoming a science, driven by the convergence of knowledge tracing, signal processing, and generative artificial intelligence (GenAI) to model student learning states and optimize education. We propose CoTutor, an AI-driven model that enhances Bayesian knowledge tracing (BKT) with signal processing techniques to improve student progress modeling and deliver adaptive feedback and strategies. Deployed as an AI copilot, CoTutor combines GenAI with adaptive learning technology. In university trials, it has demonstrated measurable improvements in learning outcomes while outperforming conventional educational tools. Our results highlight its potential for AI-driven personalization, scalability, and future opportunities for advancing privacy and ethical considerations in educational technology. Inspired by Richard Hamming’s vision of computer-aided “learning to learn,” CoTutor applies convex optimization and signal processing to automate and scale up learning analytics, while reserving pedagogical judgment for humans, ensuring that AI facilitates the process of knowledge tracing while enabling learners to uncover new insights.","PeriodicalId":13246,"journal":{"name":"IEEE Signal Processing Magazine","volume":"43 1","pages":"69-82"},"PeriodicalIF":9.6,"publicationDate":"2026-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146211293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0