Machine Learning and Artificial Intelligence at EECE

The availability of massive amounts of data and inexpensive computational capacity have recently led to unprecedented breakthroughs in machine learning methods. The machine learning and artificial intelligence research group at the Department of Electrical and Computer Engineering is a vibrant community of faculty members, graduate, and undergraduate students who are helping push the envelope of what can be accomplished using such techniques. From computer vision systems that automate security monitoring in critical infrastructure to assisted diagnosis systems, techniques developed by EECE researchers have found their way into many societally important applications. Through partnerships with industry, our energy forecasting methods based on machine learning algorithms are currently used to forecast over 20% of the gas consumption in the country. As part of Marquette's tradition of commitment to the highest standards of undergraduate education, we offer a number of courses that provide a comprehensive foundation and hands-on experience on state-of-the-art methods and technologies for the design of machine learning and artificial intelligence systems. A robust Research Experience for Undergraduate Students program funded by the National Science Foundation as well as by fellowships from the Opus College of Engineering provides opportunities for undergraduate students to be directly involved in these activities and make their own contribution to help define the future of machine learning and artificial intelligence.

 

Faculty

Dr. Cris Ababei

Dr. Priya Deshpande

Dr. Henry Medeiros

Dr. Richard J. Povinelli, P.E.

Dr. Dong Hye Ye

Courses offered in Machine Learning and Artificial Intelligence

Expand all   |   Collapse all  

Introduction to Intelligent Systems

COEN 4850  3 cr. hrs.

Provides a broad exposure to intelligent systems, including related fields such as artificial and computational intelligence. Topics include: intelligent agents, search, game playing, propositional logic and first-order predicate calculus, uncertainty, learning, communication and perception and philosophical foundations of intelligent systems. Prereq: COSC 2100, MATH 1450 and MATH 2100.

Introduction to Neural Networks and Fuzzy Systems

COEN 4860  3 cr. hrs.

Concepts of artificial neural network architectures and training algorithms, supervised and unsupervised learning, linear and non-linear neural networks, feedback neural networks, applications in scientific and engineering areas, fundamentals of fuzzy sets and fuzzy logic, fuzzy rules and inference systems, fuzzy pattern classification and clustering analysis and fuzzy control systems. Prereq: COSC 2100 and MATH 1451.

Evolutionary Computation

COEN 4870 3 cr. hrs.

Covers a set of search methods based on the Darwinian principle of survival of the fittest. The methods include genetic algorithms, evolutionary strategies and evolutionary and genetic programming, which have been successfully applied to many different problem domains including optimization, learning, control, and scheduling. Provides students with the background and knowledge to implement various evolutionary computation algorithms, discusses trade-offs between different evolutionary algorithms and other search methods, and discusses issues related to the application and performance evaluation of evolutionary algorithms. Prereq: COSC 2100, MATH 1450 and MATH 2100.

Machine Learning for Image Processing

COEN 4890 1-3 cr. hrs.

Course content is announced prior to each term. Students may enroll in the course more than once because subject matter changes. Depending upon the subject matter and the instructor, the class may be taught in traditional lecture format or as a seminar which focuses on readings from the current literature. Possible topics include advanced hardware (MPP, EPIC, VLIW), advanced software (enterprise systems, embedded software, real-time software) and advanced intelligent systems. Prereq: Cons. of instr. or Sr. stndg. 

Deep Learning Software

COEN 4690 3 cr. hrs.

Course content is announced prior to each semester. Students may enroll in the course more than once because subject matter changes. COEN design elective. Prereq: Cons. of instr.

Artificial Intelligence

EECE 6820  3 cr. hrs.

Provides a comprehensive survey of artificial intelligence. Topics include: search, logic, planning, uncertainty, learning, communication and perception, robotics and philosophical foundations of artificial intelligence. Prereq: COSC 2010, MATH 1450, MATH 2105 or equiv.

Machine Learning

EECE 6822 3 cr. hrs.

An introduction to a range of adaptive computer algorithms that learn models from data. Explores the theoretical foundations of machine learning, including computational learning theory and PAC learnability. Examples of machine learning algorithms studied include: decision trees, artificial neural networks, Bayesian learners, evolutionary algorithms and ensemble techniques. Prereq: EECE 6820 or equiv.

Neural Networks and Neural Computing

EECE 6840 3 cr. hrs.

Advanced concepts of artificial neural networks and neural computing. Mathematical modeling of neural network architectures including feed-forward and recurrent neural network models. Optimization algorithms in the neural network training. Kohonen Feature Maps (KFM), Learning Vector Quantization (LVQ) and Support Vector Machine (SVM) models. Applications include: optimization, pattern recognition and intelligent controls. Prereq: MATH 1451 or equiv.

Research Labs

Expand all   |   Collapse all  

Knowledge and Information Discovery (KID) Lab

The Knowledge and Information Discovery (KID) lab focuses on problems at the intersection of machine learning and signal processing. Graduate and undergraduate students work on problems ranging from speech recognition to electricity demand forecasting.

Learn more about the KID Lab.

Machine Learning and Image Processing

 

At Marquette’s Machine Learning and Image Processing Lab, students do research in image processing, including CT reconstruction, Organ Segmentation, Metal Artifact Reduction, Image Enhancement, Image Classification, Automatic Target Recognition, Moving Object Tracking, and Microscopic Imaging. They leverage knowledge from the state-of-the-art machine learning such as deep learning in addition to traditional statistics and signal processing

Learn more about the Machine Learning and Image Processing Lab.

Marquette Embedded Systems (MESS) Lab

 

The Marquette Embedded SystemS (MESS) Lab, located in Haggerty Hall, conducts research in the areas of design and design automation for embedded systems, multicore processors, and field programmable gate arrays (FPGAs). The outcomes of this research include hardware implementations, embedded firmware, and software programs as electronic design automation (EDA) tools to support new design methodologies or as optimization algorithms to solve specific optimization problems in these areas. The lab contains modern computers, some of which include 16 core processors and GPU devices (Tesla and Pascal architectures), and development software as well as various electronic equipment such as power sources, oscilloscopes, high accuracy current probes, microscopes, 3D printers, and reflow ovens.

Learn more about the Marquette Embedded Systems Lab.