IEEE Signal Processing Magazine最新文献

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3719401

引用次数: 0

The Geometry of LISTA [Perspectives] LISTA的几何结构[透视]

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2025.3638279

Hans van Gorp;Oisín Nolan;Yonina C. Eldar;Ruud J. G. van Sloun

{"title":"The Geometry of LISTA [Perspectives]","authors":"Hans van Gorp;Oisín Nolan;Yonina C. Eldar;Ruud J. G. van Sloun","doi":"10.1109/MSP.2025.3638279","DOIUrl":"https://doi.org/10.1109/MSP.2025.3638279","url":null,"abstract":"The celebrated iterative soft thresholding algorithm (ISTA) and its accelerated variant, fast ISTA (FISTA), are classical signal processing methods used to solve the LASSO problem [1]. This problem spans various applications, including medical imaging, direction-of-arrival estimation, astronomy, and sparse coding. Within the broader trend in signal processing of transitioning from classical model-based approaches to deep learning-based methods, the Learned ISTA (LISTA) algorithm was proposed as a way to solve the LASSO problem using deep learning while preserving the original ISTA structure [2]. LISTA learns a fast approximation to the LASSO problem by casting the weight matrices of the ISTA algorithm as learnable parameters and making them unique for each iteration, a technique now known as deep unfolding [3]. LISTA achieves superior reconstruction results in fewer iterations compared to ISTA. This is due to two main factors. First, by learning its weights, LISTA addresses potential modeling mismatches, such as imperfect knowledge of noise behavior or the forward model. Second, even with an accurately known model, the sparsifying basis might be too complex to implement efficiently using classical methods. Here, we consider a geometric interpretation to gain insight into why LISTA performs well with significantly fewer iterations (or folds) compared to ISTA. To that end, we use the fact that both models are continuous piecewise linear (CPWL) functions. Our main contributions are as follows: • We demonstrate that existing bounds on the complexity of the geometry for ISTA and LISTA are insufficient to explain their differences; their geometries must be assessed experimentally. • We introduce the concepts of expected knot density and decision density as practical metrics to evaluate the geometry of these algorithms. • We establish a lower bound on the MAP optimal decision density for sparse linear inverse problems. • We demonstrate the effect of the loss function on the geometry of LISTA. Training LISTA with an L1 norm produces fewer larger regions and a lower knot density, compared to an L2 loss, which creates many small regions and a higher knot density. • We highlight that LISTA converges faster than ISTA and, when trained with L1, produces a simpler geometry with lower knot and decision densities closer to the optimal.","PeriodicalId":13246,"journal":{"name":"IEEE Signal Processing Magazine","volume":"43 4","pages":"8-13"},"PeriodicalIF":10.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11655384","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148718974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IEEE SPS Corporate Engagement Program IEEE SPS企业参与计划

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3719402

引用次数: 0

ICIP Tampere 2026

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3719405

引用次数: 0

Demystifying CNNs: The Role of Domain Knowledge in Enhancing Interpretability and Efficiency [Lecture Notes] 揭秘cnn：领域知识在提高可解释性和效率中的作用[课堂讲稿]

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3674060

Huijun Xing;Shengxi Li;Xinyi Zhao;Harry J. Davies;Ljubiša Stanković;Danilo P. Mandic

{"title":"Demystifying CNNs: The Role of Domain Knowledge in Enhancing Interpretability and Efficiency [Lecture Notes]","authors":"Huijun Xing;Shengxi Li;Xinyi Zhao;Harry J. Davies;Ljubiša Stanković;Danilo P. Mandic","doi":"10.1109/MSP.2026.3674060","DOIUrl":"https://doi.org/10.1109/MSP.2026.3674060","url":null,"abstract":"The development of Convolutional Neural Network (CNN) architectures has traditionally relied on <italic>ad hoc</i> and <italic>brute force</i> approaches, often with limited justification for design choices. This lack of theoretical grounding has been detrimental to their broader adoption in critical domains such as healthcare, finance, and energy systems. To this end, we revisit the operation of CNNs through the lens of matched filtering, a classical signal detection technique rooted in systems science. Such conceptual insight provides a first-principles framework for understanding CNN functionality, enabling the interpretability of the canonical <italic>convolution-activation-pooling</i> pipeline, and facilitating their informed design. Moreover, this perspective supports domain-aware initialisation, suggesting ways to improve architectural efficiency and accelerate convergence. It is our hope that this Lecture Note will help establish a seamless bridge between well-established signal processing principles and the often opaque methodologies of deep learning, thereby demystifying CNNs for data analytics practitioners and for educational purposes.","PeriodicalId":13246,"journal":{"name":"IEEE Signal Processing Magazine","volume":"43 4","pages":"99-110"},"PeriodicalIF":10.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11655377","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148719004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SPS Social Media SPS社交媒体

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3717133

引用次数: 0

An AI Teaching Assistant for Motion Picture Engineering [SP Education] 电影工程人工智能教学助理

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3665423

Deirdre O’Regan;Anil C. Kokaram

{"title":"An AI Teaching Assistant for Motion Picture Engineering [SP Education]","authors":"Deirdre O’Regan;Anil C. Kokaram","doi":"10.1109/MSP.2026.3665423","DOIUrl":"https://doi.org/10.1109/MSP.2026.3665423","url":null,"abstract":"The rapid rise of LLMs over the last few years has promoted growing experimentation with LLM-driven AI tutors. However the details of implementation, as well as the benefit in a teaching environment, are still in the early days of exploration. This article addresses these issues in the context of implementation of an AI Teaching Assistant (AI-TA) using Retrieval Augmented Generation (RAG) for Trinity College Dublin?s Master?s Motion Picture Engineering (MPE) course. We provide details of our implementation (including the prompt to the LLM, and code <xref>[1]</xref>(#fn-0005)), and highlight how we designed and tuned our RAG pipeline to meet course needs. We describe our survey instruments and report on the impact of the AI-TA through a number of quantitative metrics. The scale of our experiment (43 students, 296 sessions, 1,889 queries over 7 weeks) was sufficient to have confidence in our findings. Unlike previous studies, we experimented with allowing the use of the AI-TA in open-book examinations. Statistical analysis across three exams showed no performance differences regardless of AI-TA access ( <italic>p</i>>0.05), demonstrating that thoughtfully designed assessments can maintain academic validity. Student feedback revealed that the AI-TA was beneficial (mean = 4.22/5), while students had mixed feelings about preferring it over human tutoring (mean = 2.78/5).","PeriodicalId":13246,"journal":{"name":"IEEE Signal Processing Magazine","volume":"43 4","pages":"76-84"},"PeriodicalIF":10.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11655408","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148719051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Election of Regional Directors-at-Large and Members-at-Large [Society News] 选举区域一般理事及一般会员[社会新闻]

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3714073

Ahmed Tewfik

引用次数: 0

IEEE Feedback IEEE反馈

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3717132

引用次数: 0

The IEEE Signal Processing Society’s Leading Role in Developing Standards for Computational Imaging and Sensing: Part I [SP Applications] IEEE信号处理协会在开发计算成像和传感标准中的领导作用：第一部分[SP应用]

IF 10.8 1区工程技术

IEEE Signal Processing Magazine Pub Date : 2026-07-01 Epub Date: 2026-08-13 DOI: 10.1109/MSP.2026.3678870

Miguel Heredia Conde;Corina Nafornita;Shannon-Morgan Steele;Bariscan Yonel;Venkata Veerendranadh Chebrolu;H. Brian Sequeira;Peter Vouras

{"title":"The IEEE Signal Processing Society’s Leading Role in Developing Standards for Computational Imaging and Sensing: Part I [SP Applications]","authors":"Miguel Heredia Conde;Corina Nafornita;Shannon-Morgan Steele;Bariscan Yonel;Venkata Veerendranadh Chebrolu;H. Brian Sequeira;Peter Vouras","doi":"10.1109/MSP.2026.3678870","DOIUrl":"https://doi.org/10.1109/MSP.2026.3678870","url":null,"abstract":"The practicality gap between theoretical concepts, or even lab prototypes, and commercially viable products is quite often a bridge too far. The root causes for such shortcomings may simply be that a new technology is not well suited for real-world environments or economic variables, such as cost, may be to blame. A limiting factor often overlooked in technology transfer programs is a lack of high-quality, accessible technical standards that are also definitive and relevant. Standards are essential for driving high-technology products to market that are reasonably priced for the consumer. Standards also reduce barriers to market entry for small, innovative start-up companies by incentivizing products that are interoperable between different manufacturers. The global standards development ecosystem has evolved into a very competitive landscape with significant repercussions for national job creation and economic prosperity. The voluntary consensus model has long been the predominant process for developing standards that reflect the best merit-based technology solutions to complex problems. This column is the first installment in a two-part series that highlights the preeminent role of the IEEE Signal Processing Society’s Synthetic Aperture Standards Committee (SPS-SASC) in developing standards related to computational imaging and sensing technologies and the technical leadership exercised by its active working groups.","PeriodicalId":13246,"journal":{"name":"IEEE Signal Processing Magazine","volume":"43 4","pages":"36-52"},"PeriodicalIF":10.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148718983","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