Prospect: Using Multiple Models to Understand Data [ijcai, video]
A human's ability to diagnose errors, gather data, and generate features in order to build better models is largely untapped. We hypothesize that analyzing results from multiple models can help people diagnose errors by understanding relationships among data, features, and algorithms. These relationships might otherwise be masked by the bias inherent to any individual model. We demonstrate this approach in our Prospect system, show how multiple models can be used to detect label noise and aid in generating new features, and validate our methods in a pair of experiments.
Gestalt: An IDE for Machine Learning [uist, video]
We present Gestalt, a development environment designed to support the process of applying machine learning. While traditional programming environments focus on source code, we explicitly support both code and data. Gestalt allows developers to implement a classification pipeline, analyze data as it moves through that pipeline, and easily transition between implementation and analysis. An experiment shows this significantly improves the ability of developers to find and fix bugs in machine learning systems. Our discussion of Gestalt and our experimental observations provide new insight into general-purpose support for the machine learning process.
Examining Difficulties in the Adoption of Machine Learning [chi, aaai]
As statistical machine learning algorithms and techniques continue to mature, many researchers and developers see statistical machine learning not only as a topic of expert study, but also as a tool for software development. Extensive prior work has studied software development, but little prior work has studied software developers applying statistical machine learning. This work presents two studies that provide important new insight into difficulties faced by developers and the need for development tools that better support the application of statistical machine learning.
Intelligence in Wikipedia [aaai, chi]
Although existing work has explored both information extraction and community content creation, most research has focused on these issues in isolation. In contrast, we see the greatest leverage in the synergistic pairing of these methods as two interlocking feedback cycles. We explore the potential synergy promised if these cycles can be made to accelerate each other by exploiting the same edits to advance both learning-based information extraction and community content creation.
VoiceLabel: Using Speech to Label Mobile Sensor Data [icmi]
Although existing work has explored both information extraction and community content creation, most research has focused on these issues in isolation. In contrast, we see the greatest leverage in the synergistic pairing of these methods as two interlocking feedback cycles. We explore the potential synergy promised if these cycles can be made to accelerate each other by exploiting the same edits to advance both learning-based information extraction and community content creation.
Personalizing Routes [uist]
Overly complicated directions generated by navigation services increase the cognitive load of the user, which may lead to a dangerous driving environment. We have developed a system, called MyRoute, that reduces route complexity by creating user specific routes based on a priori knowledge of familiar routes and landmarks.

In an earlier life, I was a Masters student at Stanford University focusing on Artificial Intelligence. While at Stanford, I worked in the Robot Learning Lab, the Virtual Human Interaction Lab, and with the Natural Language Group. I worked on projects dealing with sensor actuation, error visualization, turing tests, learning in virtual reality, structure from motion, and hypernym induction. I also assisted in teaching the graduate artificial intelligence course and taught a robot to play the piano.

Before Stanford, I was an undergraduate at Carnegie Mellon University double-majoring in Computer Science and Human Computer Interaction. I spent summers in the Speech Lab building and testing speech interfaces for wearable computers. As part of my senior project, I researched digital tools for online note-taking.