CMU Computer Science — Classes will not begin until September 11 2006, after the CSD Immigration Course

Machine Learning is concerned with computer programs that automatically improve their performance through experience (e.g., programs that learn to spot high-risk medical patients, recognize human faces, detect credit card fraud, and drive autonomous robots). This course covers the theory and practical algorithms for machine learning from a variety of perspectives. We cover topics such as datamining, decision tree learning, neural network learning, statistical learning methods, genetic algorithms, Bayesian learning methods, explanation-based learning, and reinforcement learning. The course covers theoretical concepts such as inductive bias, the PAC learning framework, minimum description length principle, and Occam's Razor. Short programming assignments include hands-on experiments with various learning algorithms. Typical assignments include neural network learning for face recognition, and decision tree learning from databases of credit records.

This course is targeted at graduate students who need to learn about current-day research, and how to perform current-day research, in Artificial Intelligence---the discipline of designing intelligent decision-making machines.A hallmark of recent AI conference papers, journal papers and theses has been the incorporation of ideas from outside traditional AI. Techniques from Probability, Statistics, Economics, Algorithms, Operations Research and Optimal Control are increasingly important tools for improving the intelligence and autonomy of machines, whether those machines are robots surveying Antarctica, schedulers moving billions of dollars of inventory, spacecraft deciding which experiments to perform, vehicles negotiating for lanes on the freeway, or data management systems that persistently scan for anomalies and trends. Interestingly, the overall complexity of these intelligent processes still leaves open the challenge of understanding the integration of classic AI symbolic techniques with the tools from outside traditional AI.The content of this AI course is a selected set of these tools of different types. The course will cover the ideas underlying these tools, their implementation, and how to use them or extend them in your research.

Artificial neural networks combine ideas from machine learning, statistics, and pattern recognition. They draw inspiration from, and provide simplified formalizations of, theories about the workings of the brain. This course offers an introduction to neural networks for computer scientists and engineers. Prerequisites are undergraduate calculus and linear algebra, and solid programming skills. An undergraduate course in artificial intelligence or machine learning would provide helpful background but is not required. The course provides hands-on experience with a variety of neural network architectures implemented in MATLAB, and an in-depth look at problems in pattern recognition and knowledge representation.

Topics covered include: Perceptrons, the LMS learning rule, fundamentals of pattern recognition, backpropagation learning, forward and inverse models in control theory, competitive learning, self-organizing feature maps, radial basis functions, the EM algorithm, Hopfield networks, Boltzmann machines, Helmholtz machines,general recurrent networks.

10-701 Machine Learning: 12 units
(Cross-listed as 15-781 for CS PhD students only.)

It is hard to imagine anything more fascinating than automated systems that improve their own performance. The study of learning from data is commercially and scientifically important. This course is designed to give a graduate-level student a thorough grounding in the methodologies, technologies, mathematics and algorithms currently needed by people who do research in learning and data mining or who may need to apply learning or data mining techniques to a target problem. The topics of the course draw from classical statistics, from machine learning, from data mining, from Bayesian statististics and from statistical algorithmics. Students entering the class with a pre-existing working knowledge of probability, statistics and algorithms will be at an advantage, but the class has been designed so that anyone with a strong numerate background can catch up and fully participate. Due to high demand, this class will also be offered in Spring 2005.
Please refer to this link for the most recent schedule updates.

This course is only available to CSD PhD and 5th year MS students. If you are f rom another department you must register under the number 10-701 (MLD). If you have questions send email to diane@cs

Perception, broadly defined, is the construction of a representation of the external world for purposes of thinking and acting. Although we often think of perception as the processing of inputs to the sense organs, the world conveyed by the senses is ambiguous, and cognitive and sensory systems interact to interpret it. In this course, we will examine the sensory-level mechanisms involved in perception by various sensory modalities, including vision, audition, and touch. We will learn how sensory coding interacts with top-down processing based on context and prior knowledge and how perception changes with learning and development. We will look at methods of psychophysics, neuroscience, and cognitive psychology. The goals include not only imparting basic knowledge about perception but also providing new insights into everyday experiences.

Graduate Students Only. Please email Dr. Roberta Klatzy to enroll at klatzky@cmu.edu

85-806 Seminar on Autism: 9 units

Instructor: Marcel Just
Location: Baker Hall 336B
Days/Times: W 7:00PM to 9:50PM
Autism is a disorder that affects many cognitive and social processes, sparing some facets of thought while strongly impacting others. This seminar will examine the scientific research that has illuminated the nature of autism, focusing on its cognitive and biological aspects. For example, language, perception, and theory of mind are affected in autism. The readings will include a few short books and many primary journal articles. The readings will deal primarily with autism in people whose IQs are in the normal range (high functioning autism). Seminar members will be expected to regularly enter to class discussions and make presentations based on the readings.

The seminar will examine various domains of thinking and various biological underpinnings of brain function, to converge on the most recent scientific consensus on the biological and psychological characterization of autism. There will be a special focus on brain imaging studies of autism, including both structural (MRI) imaging of brain morphology and functional (fMRI and PET) imaging of brain activation during the performance of various tasks.

Prerequisites: 85:211 or 85:213 or 85:219 or 85:355 or 85:429

85-855 Introdcution to Cognitive Neuroscience: 9 units

Instructor: Nicholas Yeung
Location: Baker Hall 340A
Days/Times: TR 1:30PM to 2:50PM
An introductory-level course for undergraduates and graduate students with no prior background in cognitive neuroscience. Note: this is not a substitute for 85-765, the cognitive neuroscience core course for CNBC students.

Please see Erin Donahoe in BH 342 E or donahoe@andrew.cmu.edu to register you.

This course deals with the science and engineering of computer vision, that is, the analysis of patterns in visual images of the world with the goal of reconstructing and understanding the objects and processes in the world that are producing them. The emphasis is on physical, mathematical, and information processing aspects of vision. Topics covered include image formation and representation, camera geometry and calibration, multi-scale analysis, segmentation, contour and region analysis, energy-based techniques, reconstruction of based on stereo, shading and motion, 3-D surface representation and projection, and analysis and recognition of objects and scenes using statistical and model-based techniques. The material is based on a recent graduate-level textbook augmented with research papers, as appropriate. The course involves considerable Matlab programming exercises.

Textbook Information:
Title: "Computer Vision: A Modern Approach"
Authors: David Forsyth and Jean Ponce
Publisher: Prentice Hall
ISBN: 0-13-085198-1

36-749 Experimental Design for Behavioral and Social Sciences: 12 units
Cross-listed as 36-309

NROSCI 2005 Cognitive Neuroscience CR HRS: 3.0 [CNBC core course]
Cross-listed as CMU 85-765

This course deals with human neuroanatomy and covers the basic structure of the central nervous system from spinal cord to cerebral cortex. Emphasis is placed on major sys tems and subsystems within the brain, and on their function al significance. The basic structure and morphology of nerve cells will be covered.

Prerequisites: NROSCI 1000 or 1003. Special Enrollment Counseling is required for registration. Students should contact Dr. Sesack for permission to register.

NROSCI/MSNBIO 2100 Cellular and Molecular Neurobiology 1: CR HRS: 4.0 [CNBC core course]
NROSCI/MSNBIO 2101 Cellular and Molecular Neurobiology 2: CR HRS: 4.0 [CNBC core course]

The Cellular and Molecular course, taught in two parts, covers protein structure and function, gene expression, neuronal development, membrane properties, the action potential, synaptic transmission, and second messenger systems, and synaptic plasticity. It consists of a lecture section and a conference section; students should attend both. The conference section is devoted to discussion of papers in the primary literature.

The 2100/2101 sequence assumes a substantial background in biology. For this reason, CNBC students not in the Neuroscience program might prefer to take the graduate section of Jon Johnson's undergraduate neurophysiology course, NROSCI 2012, in the Spring instead. The regular syllabus will be augmented with half a dozen supplementary lectures specifically for CNBC graduate students.

Nationally and Internationally recognized neuroscience researchers present scientific findings. Students meet with speakers to discuss seminar topics.

Presentations of current research by students and faculty of the Center for Neural Basis of Cognition, and by visisting researchers from other universities. Areas of cognition covered include perception, memory, language, attention, motor control and executive functions. Methodologies include single neuron recording studies, functional brain imaging studies, computational modeling studies and behavioral investigations using normal populations and idividuals with cognitive disorders.

The aim of this course is to provide a critical background in research strategies, phenomena, empirical research, and models of adult psychopathology. The course emphasis will be on etiological and pathological research, with both psychological and biological findings to be discussed. Course concentration will be on the major psychopathologies with clinical onset in adulthood, including schizophrenia, affective disorders, anxiety, addictions, and eating disorders. Conceptual and methodological issues that cross diagnostic categories will be stressed. Treatment approaches and differential diagnosis will be covered but not emphasized.

This is basic half-term module in learning and memory. It offers intensive coverage of the basic theoretical and experimental approaches to learning and memory.