Computer Science

This course gives an introduction to several subdomains of intelligent autonomous systems and robotics, and an orientation about fundamental methods and algorithms within these domains. Content covered includes three-layer architecture, Perception-Action Cycle, Robotic architectures, world models, Robot Perception, SLAM, reasoning under uncertainty, MAP-Slam, actuation, picking, placing, and reasoning and planning. 

This course examines advanced concepts including software validation and verification, the theory of testing, and advanced design patterns. The course has a strong focus on the theoretical underpinning of software design. In the labs the theory is applied with contemporary tools with concrete examples.

This course examines the basic concepts and modern software architectures on distributed and parallel computing. Topics include: computer network primitives, distributed transactions and two-phase commits, webservices, parallelism and scalability models, distributed consistency models, distributed fault-tolerance, actor and monads, Facebook photo cache, Amazon key-value stores, Google Map-reduce, Spark, and TensorFlow.

Do Google and Facebook understand us better than we do ourselves? Are we becoming lab rats every time we go online? Is the impartially designed algorithm for predicting the probability of recidivism truly fair for sentencing individuals? When big data analytics are routinely applied in our daily lives, the ability to audit the adopted algorithms becomes crucial. This course aims to build students’ big data literacy through three major areas of focus: (1) Defining what big data is; (2) Providing an overview of existing big data analytical techniques; and (3) Discussing opportunities and challenges of big data analytics in tackling social problems. 

This course introduces topics in computer and communication networks. Course topics include: network structure and architecture; reference models; stop and wait protocol; sliding window protocols; character and bit oriented protocols; virtual circuits and datagrams; routing; flow control; congestion control; local area networks; issues and principles of network interconnection; transport protocols and application layer; and examples of network protocols.

This course examines the basic elements of artificial intelligence (AI) through understanding examples from various applications and hands-on experimentation using AI software tools. In addition to covering the technical aspect of AI through such topics as search and problem solving, knowledge representation, probabilistic reasoning, machine learning, computer vision and image processing, speech and language processing, and robotics, this course will also study the historical perspective, social and ethical implications, as well as potential and limitations of AI. 

This course reviews algorithms and machine learning techniques such string pattern matching algorithms, PCA, decision tree, artificial neural networks, support vector machines, and frequent pattern mining techniques. It also reviews computational tools for algorithms and machine learning (mostly with Tensorflow, PyTorch), and surveys how these techniques are used for practical applications.

This course provides an advanced understanding of the python programming language and its main features through various applications in many fields. Students use procedural and object-oriented programming language concepts in real programs; combine programming techniques to solve problems of varying degrees of difficulty in applied fields; find and understand programming language documentation to learn new information needed to solve programming problems; and implement problem solving strategies. Course topics include input/output in Python, classes, databases management with Python, computer simulations, and agent-based modeling. Prerequisites: an introductory course on python programming or similar language (e.g. Java, C, etc.).

The extensive independent study field research paper produced by the student is both the centerpiece of the intern's professional engagement and the culmination of the academic achievements of the semester. During the preparatory session, IFE teaches the methodological guidelines and principles to which students are expected to adhere in the development of their written research. Students work individually with a research advisor from their field. The first task is to identify a topic, following guidelines established by IFE for research topic choice. The subject must be tied in a useful and complementary way to the student-intern's responsibilities, as well as to the core concerns of the host organization. The research question should be designed to draw as much as possible on resources available to the intern via the internship (data, documents, interviews, observations, seminars and the like). Students begin to focus on this project after the first 2-3 weeks on the internship. Each internship agreement signed with an organization makes explicit mention of this program requirement, and this is the culminating element of their semester. Once the topic is identified, students meet individually, as regularly as they wish, with their IFE research advisor to generate a research question from the topic, develop an outline, identify sources and research methods, and discuss drafts submitted by the student. The research advisor also helps students prepare for the oral defense of their work which takes place a month before the end of the program and the due date of the paper. The purpose of this exercise is to help students evaluate their progress and diagnose the weak points in their outline and arguments. Rather than an extraneous burden added to the intern's other duties, the field research project grows out of the internship through a useful and rewarding synergy of internship and research. The Field Study and Internship model results in well-trained student-interns fully engaged in mission-driven internships in their field, while exploring a critical problem guided by an experienced research advisor.

This course provides students an opportunity to practice symbolic logic based on mathematical fundamentals from Boolean functions and logic circuit design including assignments in Prolog language. 

Computers built from logical circuitry are a recent invention. Logic, however, has ancient roots in the attempt to distinguish sound modes of reasoning from faulty ones. It thus deals directly with language and the mind. Mathematical logic asks what an acceptable mathematical proof is, how we can justify reasoning with the infinite, etc. The formalization of mathematics through logic has clarified these questions; given mathematics a firm foundation, and, not by accident, produced a theory of computable functions, even before there were computers.

Many famous results in mathematical logic, however, are ‘negative’: demarcations of the limits of formal methods, examples of non-computability, unprovability, etc. Unsurprisingly, these negative and abstract achievements do not easily translate into practical applications.
Nevertheless, as logic structures both human reasoning and electronic computation, it can be turned into a rather nifty programming language (PROLOG) and there is an active research community applying it to cognitive science, natural languages, data mining, machine learning, artificial intelligence, fun, and more.

The goal of the course is to provide a firm grasp of some key concepts of highly abstract logic and permit them to cross the surprisingly short bridge from this idyllic realm to practical application in [room N307] reality. The logic lectures are intended to provide a theoretical vantage point. The Prolog practice enables students to represent knowledge in a program, read and understand Prolog programs, and use Prolog to solve problems.