Classes marked with an asterisk (*) count towards the PhD program in Computer Science.
This course provides an introduction to geometric algorithms and geometric data structures. Computational Geometry is a young discipline which enjoys close relations to mathematics and to various application areas such as geometric databases, molecular biology, sensor networks, visualization, geographic information systems (GIS), VLSI, robotics, computer graphics and geometric modeling. Covered topics include fundamental geometric algorithm design and analysis paradigms, geometric data structures for planar subdivisions and range searching, algorithms to computer the convex hull, Voronoi diagrams, and Delaunay triangulation, as well as selected advanced topics.
The aim of this course is to provide the student with an introduction to the main concepts and techniques playing a key role in the modern arena of artificial intelligence. In addition to covering the main topics that concern modern AI, particular attention will be devoted to its applications in several fields. Among the topics covered are, “What is an intelligent artificial agent?”, problem solving using search and constraint satisfaction, uncertainty, Bayesian networks and probabilistic inference, supervised learning, planning, sequential decision problems, as well as several additional topics.
This course has two main goals. The first one is to give a broad overview of the fundamentals of multi-agent systems (MAS). MAS are playing an increasingly important role in Artificial Intelligence as distributed resources push for highly distributed forms of intelligence.
The second aim is to provide a more in depth discussion of selected MAS topics: game theory and voting from a computational point of view. Situated at the nexus between economics and computer science, these research areas provide a perfect example of interdisciplinary cross-fertilization and mutual enrichment and lie at the core of multi-agent systems theory. The course will provide the student with an understanding of how self-interested behavior and coordination can be formally modeled and implemented in societies of artificial agents.
The aim of this course is to provide the student with an introduction to the main concepts and techniques required for collecting, processing, and deriving insight into data. Data Science is an interdisciplinary set of topics that includes everything you need to create data driven answers and solutions to specific business, scientific, or sociological questions. Topics typically covered include an introduction to one or more data collection and management systems, e.g., SOL, web scraping, and various data repositories; exploratory and statistical data analysis, e.g., bootstrapping, measures of central tendency, hypothesis testing and machine learning techniques including linear regression and clustering; data and information visualization, e.g., plotting and interactive charts using various technologies; and presentation and communication of the results of these analyses.
Over the last few decades, as we have been able to determine whole genome sequences, structural biologists have sought to determine and catalog protein structures with an increasing reliance on computational methods. Automated methods to analyze protein structure make it possible to leverage information from previously solved structures, and to interpret experimental data in a principled way. In this course, we will focus on the myriad of algorithms for analyzing numerous aspects of protein structure and protein-protein interactions.
This course provides an introduction to the fundamental concepts of machine learning and statistical pattern recognition. In addition, several examples of applications will be described. The topics covered include generative/discriminative and parametric/non-parametric supervised learning, including neural networks; unsupervised learning, including clustering, dimensionality reduction and kernel methods; learning theory, including tradeoffs, large margins and VC theory; reinforcement learning, including criteria for optimality, brute force methods, value function methods and direct policy search; feedforward/feedback adaptive control, direct/indirect adaptive control methods; and various applications.
This course studies the mathematics underlying computer security, including both public key and symmetric key cryptography, crypto-protocols and information flow. The course includes a study of the RSA encryption scheme, stream and clock ciphers, digital signatures and authentication. It also considers semantic security and analysis of secure information flow. (Same as MATH 6250.)
This course focuses on advanced techniques in the design and analysis of algorithms and illustrates how they are used in deriving a variety of now-classic results. Topics include graph algorithms, randomized algorithms, parallel computing, linear programming and approximation algorithms. (Same as MATH 6260.)
This course is an introduction to Shannon’s mathematical theory of information. It considers basic concepts such as information content, entropy and the Kullback-Leibler distance, as well as areas such as data compression and Shannon’s Source Coding Theorem, coding, prefix codes, lossless channels and their capacity, and Shannon’s Noisy Coding Theorem. Applications to various areas are also featured in the course. (Same as MATH 6280.)
This is a project-oriented course on fundamentals of software development and software engineering. Working in teams, students apply a recognized software engineering methodology, a modern programming language and software development tools (including an IDE, debugger, version control system, and testing framework) to design and implement a semester-long project – a software solution for a real-world problem.
The high goal of the course is to train students to function efficiently in a real-world software development environment. To help reach that goal, the students of the course do a lot of independent learning, teamwork, documentation and public presentation of their product and design process.
The particular technologies employed in the course may change in synchrony with changes in software engineering field, currently the focus is on engineering software-as-a-service using Ruby for programming language and Rails for web development framework.
This course is an introduction to logic and its applications in computer science. The topics covered include soundness and completeness of propositional logic, predicate logic, linear time temporal logic and branching time temporal logics and their expressive power, frameworks for software verification, Hoare triples, partial and total correctness, modal logics and agents, and binary decision diagrams.
A comprehensive introduction to the mathematics and algorithms that drive today's digital special effects, animation, and games. Designed as a hands-on course, students will gain experience in building 2D/3D interactive applications using OpenGL. Topics covered will include geometric transformations, projections, raster algorithms, 3D object models (surface and volume), visible surface algorithms, texture mapping, lighting/shading, ray-tracing, anti-aliasing, and compositing.
An introduction on how graphical representations of data can be used to aid understanding. This course details the theory and practice of designing effective information or scientific visualizations. The techniques learned in this class have wide applications to all fields in engineering and science, where due to increasing sizes and complexity data now demands effective presentation and analysis. Topics will include iso-surfacing, volume rendering, transfer functions, vector/tensor fields, topological analysis, large data visualization, and uncertainty in visualizations.
This course covers fundamental algorithm design principles and data structures, basic notions of complexity theory, as well as an advanced introduction to parallel algorithms, randomized algorithms, and approximation algorithms. Topics include: divide-and-conquer, dynamic programming, amortized analysis, graph algorithms, network flow, map reduce, and more advanced topics in approximation algorithms and randomized algorithms.
This course is designed for graduate students interested in understanding the design of autonomous intelligent agents. The course will cover fundamental notions and concepts such as uninformed and informed search, local search, constraint satisfaction and constraint-based optimization, Bayesian Networks, Markov Decision Problems and a short introduction on machine learning. Furthermore, advance topics and applications in the context of natural language processing, reasoning about time, algorithmic game theory and computational social choice will be considered as well.
This course is an introduction to computational methods in molecular biology. Topics covered include: sequence analysis and alignment, sequencing technologies, genome and metagenomic sequencing, protein structure and structure prediction, and phylogenetic analysis. No prior background in biology is assumed.
This course covers fundamental and advanced principles for designing and analyzing geometric algorithms and data structures, and their application to other disciplines. Computational Geometry is a young discipline which enjoys close relations to mathematics and to various application areas such as geometric databases, molecular biology, sensor networks, visualization, geographic information systems, VLSI, robotics, computer graphics and geometric modeling. Selected topics may include: Dynamic and kinetic data structures, geometric algorithms and data structures in two and higher dimensions, shape analysis and matching, robustness and implementation issues, geometric approximation algorithms.
This course varies from time to time, focusing on topics of interest to the faculty and students.
This course is an advanced introduction to the area of computational complexity. Topics covered include: impossibility and separability results for classical computation, interactive theorem proving and the PCP theorem, derandomization and hardness of approximation, and the quantum model of computation.
This course will cover fundamental and advanced topics in machine learning. Topics will include linear and logistic regression, Lasso, preceptrons, deep neural networks, support vector machines, kernel methods, graphical models, principal and independent component analysis and Gaussian processes. In addition to thoroughly addressing theoretical aspects, several examples will illustrate the application of the different techniques.
This course focuses on advanced machine learning techniques and their application in the domain of natural language processing (NLP). Many real world machine learning problems deal with mapping data to structured representations, and in the NLP domain the output structure represents lexical, syntactical or semantic aspects of the text. Techniques covered in this course include discriminative and generative models including Structured Perceptron and Structured SVMs, Constrained Conditional Models, Probabilistic Graphical Models, as well as Deep Learning techniques.
Reinforcement learning (RL) has found successful applications in various domains, including recommender systems, health care, energy, finance, robotics, transportation, and computer systems. Many people believe that RL is a step toward Artificial General Intelligence (AGI). This course introduces both the classic results and state-of-the-art research in RL at the graduate level. We will cover both the theoretical foundation of RL and its applications through case studies. Prerequisite: CMPS 2200 Introduction to Algorithms or equivalent or instructor approval.
The objective of the course is to introduce students to the core concepts and analytic techniques in the design and analysis of computer networks and network protocols. We will explain both how computer networks work using the Internet as the paradigm and why they work from an optimization and control perspective.
This course covers the fundamental concepts in distributed computing. The objective is to introduce students to the core notions, algorithms, and analytic tools in the design of distributed systems. Recent developments in peer-to-peer systems, cloud computing, sensor networks, etc. will be used as case studies to help students establish a firm understanding of the philosophy and pitfalls in the design of computer systems when there is no global clock and when unpredictable failures and variable latency are the norm.
This course is designed for both graduate students and advanced undergraduate students interested in understanding of both the fundamental and advanced concepts, techniques, and technologies required for collecting, processing, and deriving insight into data. Data Science is an interdisciplinary set of topics that includes everything you need to create data driven answers and solutions to specific business, scientific, or sociological questions. Topics typically covered include an introduction to one or more data collection and management systems, e.g., SQL, web scraping, and various data repositories; exploratory and statistical data analysis, e.g., bootstrapping, measures of central tendency, hypothesis testing and machine learning techniques including linear regression and clustering; data and information visualization, e.g., plotting and interactive charts using various technologies; and presentation and communication of the results of these analyses. Students should be comfortable programming in Python and familiar with the fundamentals of algorithmic analysis and computer systems.
This is a directed study course that allows a student to pursue a topic of particular interest under the direction of a computer science faculty member.
Prerequisites: Approval of the department.
This seminar course introduces students to research methods in Computer Science and to the research conducted in the department. Students will read research papers, participate in active research projects, and practice preparing and presenting research presentations. Department faculty will present on their research in order to expose students to the research projects conducted in the department. Research methods such as literature search, experiment design, technical writing, etc. will also be covered.
This course is required for all PhD students in Computer Science. The content of this course varies from semester to semester·.
In this course PhD students engage in a research project in Computer Science, under the direction of a faculty member, normally the student's faculty advisor. At the beginning of the course the scope of the project will be determined by a project proposal prepared by the student. Research will be conducted on an individual basis with the faculty advisor. The student will prepare a final report that summarize the research methodologies and the research outcomes.
The content of this course varies from semester to semester·.
This course varies from time to time, focusing on advanced topics of interest to the faculty and students.
Prerequisites: Approval of the department.
This is a directed study course that allows a graduate student to pursue a topic of particular interest under the direction of a computer science faculty member.
Prerequisites: Approval of the department.