Student Talks
Carnegie Mellon University
Learning to Learn for Small Sample Visual Recognition
Abstract: Understanding how humans and machines recognize novel visual concepts from few examples remains a fundamental challenge. Humans are remarkably able to grasp a new concept and make meaningful generalization from just few examples. By contrast, state-of-the-art machine learning techniques and visual recognition systems typically require thousands of training examples and often break down if [...]
Carnegie Mellon University
Understanding Machine Vision through Human Vision
Abstract: Recent success in machine vision has been largely driven by advanced computer vision methods, most commonly known as deep learning based methods. While we have seen tremendous performance improvements in machine visual tasks, such as object categorization and segmentation, there remain two major issues in deep learning. Firstly, deep networks have been largely unable [...]
Model Predictive Path Following for Wheeled Mobile Robots
Abstract: The navigation success of a wheeled mobile robotic mission is directly correlated to the degree of accuracy to which the robot can follow a given path. This, in turn, is largely affected by two factors: a) the environment and b) the intrinsic properties of the robot – its design, actuation mechanism etc. In the [...]
Carnegie Mellon University
Generative Models of Orbital and In Situ Data for Autonomous Science
Abstract: The mapping and characterization of planetary bodies relies on the analysis of data collected by spacecraft and orbiters. For example, the instruments carried by the Mars Reconnaissance Orbiter have been crucial in the mapping of landforms, stratigraphy, minerals, and ice of Mars. These instruments provide extensive contextual information, but factors such as sparsity, resolution, [...]
Carnegie Mellon University
Automated design, accessible fabrication, and learning-based control on cable-driven soft robots with complex shapes
The emerging field of soft robots has shown great potential to outperform their rigid counterparts due to the soft and safe nature and the capability of performing complex and compliant motions. Many are built, but the designs are conservative and limited to regular shapes. The widely-used fabrication method contains bulky pumps, tethered tubings, and silicone [...]
Carnegie Mellon University
Learning to Forecast Egocentric and Allocentric Behavior in Diverse Domains
Abstract: Reasoning about the future is fundamental to intelligence. In this work, I consider the problem of reasoning about the future actions of an intelligent agent. This poses two key questions. How can we build learning-based systems to forecast the behavior of observed agents (third-person, "allocentric forecasting")? More challenging is the question: how should we [...]
Carnegie Mellon University
Designing Interactive Systems for Community Citizen Science
Abstract: Citizen science forges partnerships between experts and citizens through collaboration and has become a trend in public participation in scientific research over the past decade. Besides this trend, public participation can also contribute to participatory democracy, which empowers citizens to advocate for their local problems. This strategy supports citizens to form a community, increase [...]
Design with Interpretability in Mind: An Alternate Ethos for Data Science
Abstract: The fields of Machine Learning and Data Science generally follow the paradigm that “the ends justify the means”, where improving predictive power of an algorithm is considered of paramount value, even when implemented at the expense of model intelligibility. While accuracy is an important performance metric, interpretability should be a major consideration for many [...]
What can this robot do? Learning Capability Models from Appearance and Experiments
As autonomous robots become increasingly multifunctional and adaptive, it becomes difficult to determine the extent of their capabilities, i.e. the tasks they can perform and their strengths and limitations at these tasks. A robot's appearance can provide cues to its physical as well as cognitive capabilities. We present an algorithm that builds on these cues [...]
Carnegie Mellon University
Visual Learning with Minimal Human Supervision
Abstract: Machine learning models have led to remarkable progress in visual recognition. A key factor driving this progress is the abundance of labeled data. Unfortunately, this reliance on lots of labeled data is also a key limitation in the rapid development and deployment of vision systems. These visual recognition systems show poor performance on concepts [...]