More Like this

1. Offline Multitask Representation Learning for Reinforcement Learning

   Ishfaq, Haque; Nguyen-Tang, Thanh; Feng, Songtao; Arora, Raman; Wang, Mengdi; Yin, Ming; Precup, Doina (December 2024, 38th Conference on Neural Information Processing Systems (NeurIPS 2024))

   We study offline multitask representation learning in reinforcement learning (RL), where a learner is provided with an offline dataset from different tasks that share a common representation and is tasked to learn the shared representation. We theoretically investigate offline multitask low-rank RL, and propose a new algorithm called MORL for offline multitask representation learning. Furthermore, we examine downstream RL in reward-free, offline and online scenarios, where a new task is introduced to the agent that shares the same representation as the upstream offline tasks. Our theoretical results demonstrate the benefits of using the learned representation from the upstream offline task instead of directly learning the representation of the low-rank model. 

   more » « less
2. Highly parallelized data-driven MPC for minimal intervention shared control

   https://doi.org/10.15607/RSS.2019.XV.008  

   Broad, A.; Murphey, T.; Argall, B. (June 2019, Robotics: science and systems)

   We present a shared control paradigm that improves a user’s ability to operate complex, dynamic systems in potentially dangerous environments without a priori knowledge of the user’s objective. In this paradigm, the role of the autonomous partner is to improve the general safety of the system without constraining the user’s ability to achieve unspecified behaviors. Our approach relies on a data-driven,model-based representation of the joint human-machine system to evaluate, in parallel, a significant number of potential inputs that the user may wish to provide. These samples are used to (1)predict the safety of the system over a receding horizon, and (2)minimize the influence of the autonomous partner. The resulting shared control algorithm maximizes the authority allocated to the human partner to improve their sense of agency, while improving safety. We evaluate the efficacy of our shared control algorithm with a human subjects study (n=20) conducted in two simulated environments: a balance bot and a race car. During the experiment, users are free to operate each system however they would like (i.e., there is no specified task) and are only asked to try to avoid unsafe regions of the state space. Using modern computational resources (i.e., GPUs) our approach is able to consider more than 10,000 potential trajectories at each time step in a control loop running at 100Hz for the balance bot and 60Hzfor the race car. The results of the study show that our shared control paradigm improves system safety without knowledge of the user’s goal, while maintaining high-levels of user satisfaction and low-levels of frustration. Our code is available online athttps://github.com/asbroad/mpmisharedcontrol. 

   more » « less
3. EyeDraw: Investigating the Perceived Effects of Shared Gaze on Remote Collaborative Drawing

   https://doi.org/10.1145/3701196  

   Ma, Eryn; Dixit, Priya; Han, Andy; Marsano, Nicholas; Sparks, Brooke; Sun, Ashley; Tiangco, Lucas; Zhou, Tongyu; Huang, Jeff; Papoutsaki, Alexandra (January 2025, Proceedings of the ACM on Human-Computer Interaction)

   Shared gaze, where collaborators can see each other's point of gaze visualized on their screen in real time, is a novel non-verbal mechanism that augments remote collaborations and increases shared awareness and common grounding. While past studies have focused on well-structured tasks and analyzed task performance and efficiency, our study explores the domain of collaborative drawing for recreational purposes and focuses on collaborators' own perceptions. We surveyed 75 users of online collaborative drawing platforms who mostly drew collaboratively for recreation and artistic growth; they reported the importance of communication but also of retaining individual space despite the collaborative setting. Informed by this and prior research on shared gaze, we evaluate collaboration by allowing two collaborators to draw synchronously on a shared canvas and share their point of gaze. We conducted a study with 24 pairs that drew collaboratively under all combinations of shared gaze and voice communication. Combining voice and shared gaze was perceived to reach the best balance between tightly coupled collaboration and parallel individual execution. Shared gaze led to higher spatial awareness and less turn-taking was observed in conditions that shared gaze was present. Surprisingly, many participants found the lack of any communication medium to afford the highest degree of divergent thinking. Our findings provide guidelines for adaptive tools that consider individual preferences as well as the nature of the task to better support remote collaborations that are open-ended and prize creativity. 

   more » « less
4. Distributed Multi-Task Learning

   Wang, Jialei; Kolar, Mladen; Srebro, Nathan (January 2016, Journal of machine learning research)

   We consider the problem of distributed multitask learning, where each machine learns a separate, but related, task. Specifically, each machine learns a linear predictor in high-dimensional space, where all tasks share the same small support. We present a communication-efficient estimator based on the debiased lasso and show that it is comparable with the optimal centralized method. 

   more » « less
5. Shared Control Decreases the Physical and Cognitive Demands of Maintaining a Secure Grip.

   Trout, M. A.; Hansen, T. C.; Olsen, C. D.; Warren, D. J.; George, J. A. (August 2022, Proceedings of the Myoelectric Controls and Upper Limb Prosthetics Symposium.)

   Upper-limb amputees commonly cite difficulty of control as one of the main reasons why they abandon their prostheses. Combining myoelectric control with autonomous sensor-based control could improve prosthesis control. However, the cognitive and physical impact of shared control and semi-autonomous systems on users has yet to be fully explored. In this study we introduce a novel shared-control algorithm that blends proportional position control predicted from electromyography (EMG) with proportional position control predicted from an autonomous machine using infrared sensors embedded in the prosthetic hand’s fingers to detect the distance to objects. The user’s EMG control is damped in proportion to the machine’s prediction of an object’s position in relation to a given finger. The shared-control algorithm was validated using three intact individuals completing a holding task where they attempted to hold an object for as long as possible without dropping it. Shared control resulted in fewer object drops, 32% less cognitive demand, and 49% less physical effort (measured by EMG) relative to the participant’s EMG control alone. These results indicate that shared control can reduce the physiological burdens on the user as well as increase prosthetic control. 

   more » « less