Lorem ipsum Dolor Sit Amet, Consectetur Adipiscing

Vestibulum eget magna vitae lacus tristique vulputate

Lorem ipsum Dolor Sit Amet, Consectetur Adipiscing Elit

The Machine INtelligence and Data Science research group is a part of the University of Southern California’s Information Sciences Institute, located in sunny Marina del Rey, CA. We deploy and develop machine learning algorithms to analyze data from a variety of applications, including social science, cybersecurity, and biomedical domains.

We are currently hiring for several postdocs and internship positions!
Please see our list of openings here

See what’s happening in the world right now

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.

Artificial Intelligence at ISI

PROJECTS

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

Big Mechanism

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod

SAFE

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod

EFFECT

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod

Next Gen Social Sci

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod

PEOPLE

Lorem ipsum dolor sit amet consectetur.

Faculty

Aram Galstyan

Research Associate Professor

Kristina Lerman

Research Associate Professor

Emilio Ferrara

Research Assistant Professor

Greg Ver Steeg

Research Assistant Professor


Researchers

Fred Morstatter

Computer Scientist

Tozammel Hossain

Postdoctoral Associate

Homa Hosseinmardi

Postdoctoral Associate

Anna Sapienza

Postdoctoral Associate



PUBLICATIONS

Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Machine Learning

Our group uses machine learning across a variety of applications, but recently, a main thrust of theoretical work has revolved around developing information theoretic methods.

Mutual information is among the most powerful and general measures of the relationship between random variables, but it’s effectiveness in machine learning has been limited by -

a) The difficulty of measuring entropy and mutual information in high dimensions (see UAI-15, AISTATS-15, NIPS-16)

b) Frequent misuse of (pairwise) mutual information (e.g. ICML-14).

It can be difficult to even define meaningful information measures for the interaction of variables in high dimensions.

We have taken steps toward formalizing the use of multi-variate mutual information objectives for unsupervised learning. The method of Correlation Explanation (AISTATS-15, NIPS-14) provides an information-theoretic foundation for modularly and hierarchically decomposing information in complex systems. Information may be extracted incrementally using the “information sieve” method (ICML-16), and its extension for continuous variables.

These methods have found applications in gene expression (interesting podcast and article about this work), brain imaging, text analysis, and psychometrics.

Code for Correlation Explanation and the Information Sieve can be found here.

Computational Social Science

Our society is becoming increasingly dependent on various technological networks, particularly in the realm of social media. As these networks evolve and grow in complexity, their dynamical behavior is becoming difficult to understand and predict. We design machine-learning frameworks to model and predict individual , characterize information diffusion, predict crime and abuse, and ultimately understand human behavior. We hope that insights generated from this data can be used to support informed decision making and implementation in the social and political sciences.

Emilio’s site, including blog posts with paper summaries, is a great resource for this line of work. Also, we are hiring postdocs! See our Join Us page and contact us by email if interested.

Biomedical Applications

While our group is always looking for interesting data sets, we have found the machine learning in the biomedical domain to be particularly fruitful. Several paths of inquiry have included developing more advanced network models of brain connectivity, extracting biomolecular interaction knowledge from research texts, and using recurrent neural networks to find meaningful patterns for diagnosis in clinical time series data.

Other links of interest include: Big Mechanism project description

Cancer in the Time of Algorithms for blog post, news article, and podcast about Greg Ver Steeg and Shirley Pepke’s inspiring work identifying signals in gene expression data for patients with ovarian cancer, a disease Pepke herself is fighting.

Dan presents his work, “A Continuous Model of Cortical Connectivity” , for which he won the Young Scientist award at MICCAI!

The Machine Learning in Health Care symposium, for which Dave Kale is an organizer.

Forecasting

Real-world events can often be viewed as salient outcomes of interactions between sets of latent spatiotemporal processes that reflect behaviors of individuals and of populations, along with the evolution of environmental factors. We seek to develop advanced, automated methods of consolidating the vast amount of available data into predictions of meaningful events. This involves integrating information from a range of sources, modeling their interactions over time, and making inferences about significant patterns and/or anomalies.

Two examples of projects in this domain include SAFE, where we are developing a suite of methods for forecasting geopolitical events, and EFFECT, where we are developing an end-to-end system for detecting cyber security vulnerabilities and predicting the timing and targets of attacks.

CODE

Lorem ipsum dolor sit amet, consectetur adipiscing elit.

"Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum."

JOIN US

Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Postdoctoral Positions

We are actively hiring postdocs for work in Computational Social Science! Please see the adjacent flier and reach out by email if interested.


Internships

We are also actively looking for summer interns to join our group. Please see our How to Apply and FAQ pages for more information. The following positions are in our group:


Prospective PhD Students

Please apply through the USC Computer Science department process described here. If you mention a professor from our group in your application, we will read and consider your application.