6 accepted papers
The exponential growth in demand for GPU computing resources has created an urgent need for automated CUDA optimization strategies. While recent advances in LLMs show promise for code generation, current state-of-the-art models achieve low success rates in improving CUDA speed. In this paper, we in…
Fictitious play is an algorithm for computing Nash equilibria of matrix games. Recently, machine learning variants of fictitious play have been successfully applied to complicated real-world games. This paper presents a simple modification of fictitious play which is a strict improvement over the or…
Manipulation involves a broad spectrum of skills, e.g., polishing, peeling, flipping, screwing, etc., requiring complex and delicate control over both force and position. This paper aims at designing an optimal haptic interface for providing a robot with direct demonstrations of human's innate intel…
This paper presents the ongoing work towards enabling humanoid robots to achieve dynamic behaviors comparable to humans. By using the concept of Divergent Component of Motion (DCM) first in introduced in [1], the present paper permits operator and robot balance synchronization using the mutual dynam
This paper introduces a novel Balance Feedback Interface (BFI) that addresses the problem of bilateral feedback for teleoperation of humanoid robots. With this new device we expect to enhance robot's high force manipulation performance to a level comparable to humans by dynamically synchronizing mas…
This paper introduces the phenomenon of directional efficiency, a property of spur gear transmissions in which the coefficient of torque dependent efficiency is function of the power flow direction; the efficiency in the step-up direction differs from that of step-down. Understanding the directional…