High Efficiency
High efficiency in various computational domains is a central research theme, aiming to minimize resource consumption (time, memory, energy) while maintaining or improving performance. Current efforts focus on developing novel algorithms and architectures, such as optimized Thompson sampling for reinforcement learning, sparse attention mechanisms for transformers, and efficient model compression techniques, to achieve this goal across diverse applications including natural language processing, computer vision, and robotics. These advancements are crucial for deploying complex AI models on resource-constrained devices and for accelerating scientific discovery in data-intensive fields.
Papers
WOLONet: Wave Outlooker for Efficient and High Fidelity Speech Synthesis
Yi Wang, Yi Si
Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification
Massimiliano Patacchiola, John Bronskill, Aliaksandra Shysheya, Katja Hofmann, Sebastian Nowozin, Richard E. Turner
Efficient and Flexible Sublabel-Accurate Energy Minimization
Zhakshylyk Nurlanov, Daniel Cremers, Florian Bernard
Merak: An Efficient Distributed DNN Training Framework with Automated 3D Parallelism for Giant Foundation Models
Zhiquan Lai, Shengwei Li, Xudong Tang, Keshi Ge, Weijie Liu, Yabo Duan, Linbo Qiao, Dongsheng Li
Convex Hull Prediction for Adaptive Video Streaming by Recurrent Learning
Somdyuti Paul, Andrey Norkin, Alan C. Bovik