A major may elect to receive a B.S. in Cognitive Science with specialization in Machine Learning and Neural Computation (Major code: CG35). This area of specialization is intended for majors interested in computational and mathematical approaches to modeling cognition or building cognitive systems, theoretical neuroscience, as well as software engineering and data science. Allowed electives include advanced courses in neural networks, artificial intelligence, and computer science. The following requirements are effective for incoming students after Fa16:
LOWER DIVISION REQUIREMENTS (11 courses, 44 units or 10 courses, 40 units)
Math (chosen from the following):
UPPER DIVISION REQUIREMENTS (12 courses, 48 units)
Core (6 courses):
Electives (6 courses):
NOTE: Courses for the major must be taken for a letter grade with the exception of 195, 198, and 199 which are only offered on a P/NP basis; a minimum grade of C- is required for all courses; Complete most of your CORE courses during your junior year, if possible; At least half of all electives taken must be Cogs Courses; others can be chosen from the list of approved electives or petitioned through the department.
MACHINE LEARNING AND NEURAL COMPUTATION FACULTY
Virginia de Sa. Assistant Professor, CSB 164, 858-822-5095, firstname.lastname@example.org, website. Research: We use computational modeling, psychophysics studies, and machine learning to learn more about visual and multi-sensory perception.
Eran Mukamel. Assistant Professor, email@example.com, website. Research: Computational analysis of large-scale neural data, electrophysiology of sleep and general anesthesia, computational epigenomics in brain cells.
Zhuowen Tu. Associate Professor, SSRB 250, 858-822-0908, firstname.lastname@example.org, website. Research: Computer vision, machine learning, deep learning, neural computation, neuro imaging.
Bradley Voytek. Assistant Professor, CSB 169, 858-534-0002, email@example.com, website. Oscillatory network communication, automated science, data-mining, aging, attention, working memory, cognitive brain-computer interfaces, brain/cognition/society interactions.
Angela J. Yu. Associate Professor, CSB 157, 858-822-3317, firstname.lastname@example.org, website. Research: Decision making, attention, active vision, learning, neuromodulation, Bayesian modeling, control theory.
FACULTY WITH RELATED RESEARCH
Philip Guo. Assistant Professor, CSB 129, email@example.com, website. Research: Human-computer interaction, design, online learning, computing education, programmer productivity.
Ayse P. Saygin. Associate Professor, SSRB 202-20, 858-822-1994, firstname.lastname@example.org, website. Research: Cognitive neuroscience, neuropsychology, neuroimaging, visual perception, attention, multisensory integration, biological motion, social neuroscience, language comprehension, human-machine interaction, social robotics.
Terrence J. Sejnowski. Adjunct Professor, CNL/Salk, 858-453-4100 Ext. 1611, email@example.com, website. Research: Computational neurobiology; the representation, transformation, and storage of information in the nervous system.
RECOMMENDED MACHINE LEARNING AND NEURAL COMPUTATION COURSES
COGS 8. Hands-on Computing (4)
Introductory-level course that will give students insight into the fundamental concepts of algorithmic thinking and design. The course will provide the students with first-person, hands-on experience programming a web crawler and simple physical robots.
COGS 108. Data Science in Practice (4)
Data science is multidisciplinary, covering computer science, statistics, cognitive science and psychology, data visualization, artificial intelligence, and machine learning, among others. This course teaches critical skills needed to pursue a data science career using hands-on programming and experimental challenges. Prerequisites: Cognitive Science 18 or CSE 7 or CSE 8A or CSE 11.
COGS 109. Modeling and Data Analysis (4)
Exposure to the basic computational methods useful throughout cognitive science. Computing basic statistics, modeling learning individuals, evolving populations, communicating agents, and corpus-based linguistics will be considered. Prerequisites: Cognitive Science 14B and Mathematics 20F and CSE 7 or CSE 11 or CSE 8A or consent of instructor.
COGS 118A. Intro to Machine Learning I (4)
This course is one part of a two-course foundation that forms a rigorous introduction to machine learning and computational modeling of biological intelligence. Natural Computation I and II are independent courses that may be taken in either order. Topics in Natural Computation I may include Bayesian inference, regression, graphical models, sampling, hidden Markov model, decision theory, information theory, reinforcement learning, and some application areas. Prerequisites: CSE 11 or CSE 8B and Mathematics 20F and Mathematics 20E and Mathematics 180A, or consent of instructor.
COGS 118B. Intro to Machine Learning II (4)
This course is one part of a two-course foundation that forms a rigorous introduction to machine learning and computational modeling of biological intelligence. Natural Computation I and II are independent courses that may be taken in either order. Topics in Natural Computation II may include: maximum likelihood estimation, bayesian parameter estimation, clustering, self-organized maps, principal component analysis, neural networks, support vector machines, and some application areas. Prerequisites: CSE 11 or CSE 8B and Mathematics 20F and Mathematics H20E and Mathematics 180A or consent of instructor.
COGS 118C. Neural Signal Processing (4)
This course will cover theoretical foundations and practical applications of signal processing to neural data. Topics include EEG/field potential methods (filtering, Fourier (spectral) analysis, coherence) and spike train analysis (reverse correlation, spike sorting, multielectrode recordings). Some applications to neural imaging (optical microscopy, fMRI) data will also be discussed. Prerequisites: Mathematics 18 or Mathematics 20F or Mathematics 31AH and Cognitive Science 14B or Psychology 60 and Cognitive Science 108 or Cognitive Science 109.
COGS 118D. Mathematical Statistics for Behavioral Data Analysis (4)
Statistical methods for analyzing behavioral data. A mathematically sophisticated course covering both classical and Bayesian statistical methods for estimation, hypothesis testing, regression, and model comparison. Emphasis on both mathematical understanding of statistical methods as well as common applications. Prerequisites: Cognitive Science 108 or Cognitive Science 109 or Mathematics 11 and Mathematics 18 or Mathematics 20F or Mathematics 31AH and Mathematics180A or consent of instructor.
COGS 180. Neural Coding in Sensory Systems (4)
This course covers recent advances in the understanding of common neural mechanisms and computational principles underlying the brain’s ability to process multiple sources of sensory information—vision, audition, olfaction, touch, and equilibrioception—and translate them into actions. Prerequisites: Cognitive Science 1, Cognitive Science 14B, Cognitive Science 101A, and Cognitive Science 109.
COGS 181. Neural Networks and Deep Learning (4)
This course will cover the basics about neural networks, as well as recent developments in deep learning including deep belief nets, convolutional neural networks, recurrent neural networks, long-short term memory, and reinforcement learning. We will study details of the deep learning architectures with a focus on learning end-to-end models for these tasks, particularly image classification. Pre-req: (MATH18 or MATH20F or MATH31AH) and (ECE109 or MATH180A) and (COGS108 or COGS109 or CSE11)
COGS 185. Advanced Machine Learning Methods (4)
This course is an advanced seminar and project course that follows the Natural Computation courses. Advanced and new machine learning methods will be discussed and used. Prerequisites: Cognitive Science 118B or Cognitive Science 118A.
COGS 188. AI Algorithm and Social Language (4)
This class will cover latest machine learning and text analysis algorithm. Such algorithms have become important in the area of data collected from the Internet and for the analysis for social network activities such as Twitter posts, E-mails, blogs, etc. Prerequisites: Cognitive Science 109 or Cognitive Science 118A or Cognitive Science 118B.
COGS 189. Brain Computer Interfaces (4)
This course will discuss signal processing, pattern recognition algorithms, and human-computer interaction issues in EEG-based brain-computer interfaces. Other types of brain-computer interfaces will also be discussed. Prerequisites: Cognitive Science 118B or Cognitive Science 118A or Cognitive Science 109.