Portrait picture
Junghoo "John" Cho
Computer Science

Email: cho@cs.ucla.edu
Phone: (310) 571-8240
Office: Boelter Hall 3531H
Office hour: Tue 2:30-3:30 PM

Quick Links:  Research | Publications | Awards | Teaching

Biographical Sketch

Junghoo Cho is a professor in the Department of Computer Science at University of California, Los Angeles. He received a Ph.D. degree in Computer Science from Stanford University and a B.S. degree in physics from Seoul National University. His research interest is in the theory and practice of learning, particularly in the area of language acquisition and understanding. He is a recipient of the 10-Year Best-Paper Award at VLDB 2010, NSF CAREER Award, IBM Faculty Award, Okawa Research Award, Northrop Grunmann Excellence in Teaching Award and Dr. Stevenson Faculty-in-Residence Award.

Research Interests

My research agenda is understanding and replicating the self-learning ability of humans. I work on this research agenda because I am deeply baffled by the following question: when we are born, how are we able to learn and make sense of the world without any "explicit teaching" by others? This ability looks almost magical to me, but I am also convinced that our brain is nothing more than a computational machine that happens to be implemented using a network of neurons. Given this conviction, I feel that there must be "computational algorithms" that implement the human's learning capabilities. My research goal, therefore, is to discover the mathematical principles and the computational algorithms that lie behind the human's amazing learning capacity.

Any scientific endeavor requires data from which we get inspiration and with which we validate our hypotheses. The particular domain data that I currently work with is human language. Therefore, the current goal of my research may be summarized as "can we develop algorithms that can automatically discover the grammatical structures and the semantic relationship embedded in our language?"

As toolsets to investigate this problem, I use linear algebra, tensor analysis, signal processing, functional analysis, probabilistic and bayesian inference, information theory, and (discrete) differential geometry. As I explore this domain, I learn that topics that I thought were completely unrelated are in fact connected at a deeper layer. For example, until a few years ago, I did not know that the graph partitioning problem is related to identifying the low frequency components of its "Fourier transform", which are obtained as the eigenfunctions of the Laplace operators of the manifold in which the data reside. These are probably the things that the experts in the field already know very well, but how would I have known these when my original background is databases? And whenever I learn these deeper connections, it is so satisfying personally, even if it might be something well known among experts.

In general, I find that, again and again, at the core of everything, our world looks so simple, elegant, and tightly connected. How can this simple structure lead to the amazingly rich world that we experience every day? This is the never-ending source of mystery and wonder that drives me to work on these problems.

Recent Professional Activities



Current Ph.D. students

Past PhD Students


I teach the following classes regularly at UCLA.

© Junghoo "John" Cho