A partial listing of courses offered at UT Austin that are relevant to a computational or theoretical neuroscience focus.
Computational Neuroscience and Neural Networks (NEU 337 / EE 385V / PHY 392T)
This graduate class (open to upper-division undergrads with permission of the instructor) is cross-listed between Neuroscience, Physics, Mathematics, Computer Science, and Electrical Engineering. The goal is to introduce the theory of memory, sensory processing, and learning in the context of the brain as well as artificial neural networks. The course is organized in three parts: Recurrent neural networks for memory and integration; Supervised and unsupervised learning in recurrent and feedforward networks; Sensory systems and topics at the research frontier. The class consists of lectures, student-led presentation of primary research articles, and homework that involves programming and pen-and-paper work. The goal is to bring students from the physical sciences, mathematics, engineering, and neuroscience to the point where they could begin to engage in research in computational neuroscience or related areas if they wish. (3 lecture hours/week for one semester.)
Neural Networks (NEU 394P, CS394N)
The main goals of the class are to (1) obtain an overview of current state of the art in the field, (2) carry out a substantial research project, and (3) get practice in research skills such as conducting a literature study, putting together a research and a conference talk, and writing a research paper. The course is organized so that selecting and completing a research project should be as easy as possible.
Neural Systems (NEU 330)
Introduction to the nervous system with an emphasis on brain organization, neuron physiology, perceptual systems, and motor systems. Intended for neuroscience majors and those considering neuroscience as a major. Three lecture hours a week for one semester.
Vision Systems (NEU/PSY 380E)
This course presents an introduction to the physiology, psychophysics and computational aspects of vision. A fundamental premise of the course is that theoretical and applied research in computational vision provides insights that are useful for understanding biological vision. Conversely, physiological and psychophysical studies of human and other biological visual systems provide important clues for the development of computational vision systems. Thus, the course will proceed by analyzing biological vision systems from an engineering/information-theoretic perspective. Although the exact content of the course will vary from year to year, the topics covered will include a large subset of the following: extraction and representation of visual information in biological and machine systems; natural scene statistics, computational theories of vision; anatomy, physiology and psychophysics of the primate (human) visual system; image processing; pattern detection and identification; stereo vision; motion perception; surface and object perception; color vision. The course will include reviews of linear systems analysis, information theory, Bayesian statistical decision theory and other relevant mathematical topics.
Advanced Topics in Systems Neuroscience: Visual Processing (PSY 394U)
Much of our understanding of higher brain function comes from studies of the visual system. In this course, we will read and discuss a mixture of recent and classical papers in visual neuroscience. We will consider two central questions in visual neuroscience: First, “How does the brain represent visual information?”; and second, “How does this representation support perceptual experience and visually-guided action?”. Students will be required to read all the papers, and will be assigned several papers to present during the semester. The grade will be based on the presentations and class participation.
The Computational Brain (CS 378)
One of our major scientific challenges of the century is to understand the functioning of the human brain. Computational models play a vital role in a complete picture of brain function, particularly at modeling more macroscopic structures that more directly relate to our everyday behavior. The goal of this course is to describe computational models of intelligent behavior and how they relate to structures in the brain. Students should have a mathematical background sufficient to grasp the ideas behind learning algorithms. This would include calculus and some linear algebra. Students will write five short evaluations of scientific papers in computational brain science. Students will also review one popular science book in the area. In addition one group lab experiment will be done and its outcomes written up in a report and presented orally.
Quantitative Methods in Neuroscience (NEU 466M)
This upper-division undergraduate course (also open to graduate students) provides a broad introduction to basic mathematical and computational tools for a quantitative analysis of neural systems. Integrated lectures, programming sessions, and homework sets will introduce techniques and help us learn to apply them. We will cover a range of topics, including neural encoding and decoding, population codes, filtering, correlation, convolution, spike-triggered averaging (reverse correlation), deconvolution, and dimensionality reduction, clustering, and spike-sorting through principal components analysis, as well as some probability and Bayesian inference, as used in neuroscience. The goal is to help develop a level of intuitive and practical comfort with quantitative methods and visualization of complex data.
Machine Learning (CS 391L)
This graduate-level computer science course covers computing systems that automatically improve their performance with experience, including various approaches to inductive classification such as version space, decision tree, rule-based, neural network, Bayesian, and instance-based methods; as well as computational learning theory, explanation-based learning, and knowledge refinement.
Neural Networks (CS 394N)
This graduate-level course covers biological information processing; architectures and algorithms for supervised learning, self-organization, reinforcement learning, and neuro-evolution; theoretical analysis; hardware implementations and simulators; applications in engineering, artificial intelligence, and cognitive science.
Visual Recognition (CS 381V)
Subjects in this graduate-level computer science class include fundamental representations, learning approaches, matching-based algorithms, human activity models for video, and large-scale recognition.