About
Research & Development
DALI-PD: Accelerating ML in Chip Design with Synthetic Layout Heatmaps
TLDR: DALI-PD is a new framework that uses diffusion models to quickly generate diverse and realistic synthetic layout heatmaps for machine learning in physical...
Research & Development
Enhancing GUI Agent Security: A Layer-wise Approach to Defend Against Pop-up Attacks
TLDR: A new method called LaSM (Layer-wise Scaling Mechanism) improves the robustness of GUI agents against malicious pop-up attacks. It works by selectively amplifying...
Research & Development
FedGSCA: A New Framework for Robust Medical AI Collaboration Amidst Noisy Data
TLDR: FedGSCA is a novel federated learning framework designed to improve medical image classification by effectively handling label noise and class imbalance. It uses...
Research & Development
Accelerating Unsupervised Time-Series Segmentation with Random Fourier Features
TLDR: A new method, RFF-GP-HSMM, is proposed to make unsupervised time-series segmentation faster and more scalable. It addresses the high computational cost of the...
Research & Development
GeoHopNet: Advancing Drone Site Selection with AI in Urban Environments
TLDR: GeoHopNet is a new deep reinforcement learning framework designed to optimize dynamic site selection for UAV landing points and supply stations in urban...
Research & Development
A Simple Breakthrough for Continual AI Learning: Balancing Adaptability and Memory
TLDR: RDBP is a new, simple method for continual learning in AI, combining ReLUDown (an activation modification for plasticity) and Decreasing Backpropagation (a gradient...