Machine Learning
Learning the MMSE Channel Estimator
Accurate channel estimation is a major challenge in the next generation of wireless communication networks. To fully exploit setups with many antennas, estimation errors must be kept small. This can be achieved by exploiting the structure inherent in the channel vectors. For example, line-of-sight paths result in highly correlated channel coefficients.
Graph Neural Networks
Filtering is the fundamental operation upon which the field of signal processing is built. Loosely speaking, filtering is a mapping between signals, typically used to extract useful information (output signal) from data (input signal). Arguably, the most popular type of filter is the linear and shift-invariant (i.e. independent of the starting point of the signal) filter, which can be computed efficiently by leveraging the convolution operation.
Deep Learning on Graphs: History, Successes, Challenges, and Next Steps
Deep learning on graphs, also known as Geometric deep learning (GDL) [1], Graph representation learning (GRL), or relational inductive biases, has recently become one of the hottest topics in machine learning. While early works on graph learning go back at least a decade [2], if not two [3], it is undoubtedly the past few years’ progress that has taken these methods from a niche into the spotlight of the Machine Learning (ML) community.
Machine Learning for Readiness and Performance Improvement
Military operations and training present a broad variety of demanding physical tasks which may impact the Warfighter physical performance and health…
Read moreWing-Kin (Ken) Ma (The Chinese University of Hong Kong)
Lecture Date: November 7, 2018
Chapter:Tokyo/Fukuoka/Hiroshima/ Nagoya/<br />Sapporo/Shikoku/ Shin-Etsu Joint Chapter
Chapter Chair: Shoji Makino
Topic: Hyperspectral Unmixing: Insights and Beyond
Wing-Kin (Ken) Ma (The Chinese University of Hong Kong)
Lecture Date: June 1 & 7, 2018
Chapter: France
Chapter Chair: William Puech
Topic: (1) Hyperspectral Unmixing in Remote Sensing: Learn the
Wisdom There and Go Beyond (Machine Learning Included)
(2) MIMO Transceiver Designs and Optimization: Beyond Beamforming and
Perfect Channel Information
Wing-Kin (Ken) Ma (The Chinese University of Hong Kong)
Lecture Date: June 5, 2018
Chapter: Benelux
Chapter Chair: Francois Horlin
Topic: Hyperspectral Unmixing in Remote Sensing: Learn the
Wisdom There and Go Beyond (Machine Learning Included)
Jan Allebach (Purdue University, USA)
Lecture Date: May 1, 2018
Chapter: Santa Clara Valley
Chapter Chair: Pavel Tcherniaev
Topic: Small, Medium, and Big Data: Application of Machine
Learning Methods to the Solution of
Real-World Imaging and Printing Problems
MLSP TC
Scope
The Machine Learning for Signal Processing Technical Committee (MLSP TC) is at the interface between theory and application, developing novel theoretically-inspired methodologies targeting both longstanding and emergent signal processing applications. Central to MLSP is on-line/adaptive nonlinear signal processing and data-driven learning methodologies. Since application domains provide unique problem constraints/assumptions and thus motivate and drive signal processing advances, it is only natural that MLSP research has a broad application base. MLSP thus encompasses new theoretical frameworks for statistical signal processing (e.g. machine learning-based and information-theoretic signal processing), new and emerging paradigms in statistical signal processing (e.g. independent component analysis (ICA), kernel-based methods, cognitive signal processing) and novel developments in these areas specialized to the processing of a variety of signals, including audio, speech, image, multispectral, industrial, biomedical, and genomic signals. The MLSP TC is focused on fostering research in these areas, the application of these techniques, and in educating the technical community about research developments in these areas.
Highlights From the MLSP TC
The huge success of this wave of artificial intelligence (AI) has primarily been driven by machine learning, which provides the essential tools for analyzing signals and data that are ubiquitously available today. The MLSP TC aims at fostering novel machine learning methodologies for both longstanding and emergent signal processing applications of a broad range.
Check out this spotlight article to learn more about the MLSP TC's recent activities.
Seeking New Members
We are currently seeking nominations for new members of the MLSP TC. All candidates must be current IEEE and SPS members. They can be self-nominated or nominated by Technical Committee members. Members will serve a term of three years. At the end of the first term, current Technical Committee members may also be nominated for a second consecutive term. Past Technical Committee members are eligible to be nominated for an additional term, but at least a three-year gap in service is required. New members must be willing to review papers within the area of the Technical Committee, which are submitted to the Society’s conferences, review papers for workshops owned or co-owned by the Technical Committee, serve in the subcommittees established by the Technical Committee, and perform other duties as assigned.
The deadline for submitting applications is September 19, 2025.
The Nomination and Self-Nomination forms can be found here:
Either of these forms should be submitted together with a CV (both files in PDF) directly to the Member Nomination & Election Subcommittee and TC Chairs.
