Andrew Arnold's

Research  •  CV  •  Java

         

MOTTO:

"Man, I love science!"    – Beakman's World

ABSTRACT:

I am currently a Ph.D. student under the supervision of William W. Cohen in the Machine Learning Department within Carnegie Mellon University's School of Computer Science.

My research is generally concerned with machine learning and data mining, with an underlying interest in producing interpretable and reusable views of data, features and models. To this end, I am particularly interested in transfer learning with an emphasis on domain adaptation.

I am currently working on the SLIF system for mining text and images together. The text (captions, abstracts and main text from biomedical journal articles) directly and indirectly refers to the images, while the images depict entities (cells, proteins, interactions) that are described in the text. The combination of these two expressions (text and images) of the same underlying concept (the experiment being performed) into new features, jointly describing both the text and images, would be closer to representing the actual object a user would be interested in, rather than disjoint features of text and image alone. A related problem is that of transfer learning. In this case, we use models and named entity extractors trained on one type of data (abstract text, for instance) and adapt them to be applied to a related, but distinct type of data (caption text). The intuition is that it is easier to learn a certain concept once a related concept has already been mastered.

I have also been lucky to pursue related work outside of school during summer internships. While working with Hang Li and Tie-Yan Liu in the Web Search and Mining group at Microsoft Research Asia we developed novel semi-supervised and transfer learning based methods for improving internet search through query-dependent ranking. The idea behind this work is that, regardless of the specific topic users are interested in, there are common features linking certain types of queries together. For instance, users searching for either a person or company name might both be most interested in the corresponding home page (a navigational query), while searchers for a disease or country name might be more interested in authoritative sources of information about these topics (informational queries). By modeling and leveraging these distributions of types of queries we can better decide what, exactly, users want and deliver that to them.

Relatedly, while in the Data Analytics group at IBM Research Watson, I worked with Naoki Abe and Yan Liu on methods for learning causal models from temporally ordered data. We felt that the interpretability offered by a causal model was quite valuable for the end user in understanding the process being studied. This type of understanding is an essential component of the scientific process since it leads the researcher to an idea of what experiment to perform next. An accurate predictive model, without interpretation, provides little insight as to what direction is best to pursue. This was also the motivation behind my work with Richard Scheines and Joseph E. Beck on discovering predictive, semantically and scientifically interpretable high-level features as functions of raw, event level data.

I did my undergraduate work in the Intrusion Detection System Group within the Computer Science Department of Columbia University, under the supervision of Professor Salvatore J. Stolfo and Eleazar Eskin. My work there dealt with applying kernel methods and support vector machines to the problem of clustering data (binaries, system calls, network packets, etc.) in order to identify possible attacks.

In my spare time I work on applying machine learning techniques towards opponent modeling for Texas Hold 'em poker and Tic-Tac-Toe.

PAPERS:

INVITED ARTICLES:

REPORTS:

TALKS: