About
Research & Development
Accurate Transit Times: A New Approach to Deep Learning for Bus and Tram Predictions
TLDR: A new deep learning method, NSATP, improves bus and tram arrival time predictions by addressing "over-stationarization." While data normalization helps predictability, it can...
Research & Development
Optimizing Industrial Scheduling: A Novel DRL Environment for Flexible Job-Shops
TLDR: This paper introduces an efficient deep reinforcement learning (DRL) environment for the Flexible Job-shop Scheduling Problem (FJSP), a complex optimization challenge. Unlike previous...
Research & Development
MEGS 2: A Memory-Efficient Framework for 3D Gaussian Splatting
TLDR: MEGS 2 is a new framework that significantly reduces the memory footprint of 3D Gaussian Splatting (3DGS) for real-time rendering on various devices,...
Research & Development
Single-Bit Neural Networks Achieve High Performance with Minimal Memory
TLDR: A new research paper introduces Binary Normalized Neural Networks, a novel approach where all neural network parameters (weights and biases) are restricted to...
Research & Development
Simplifying Decision-Making Under Uncertainty: A Minimalist Bayesian Approach
TLDR: A new research paper introduces a 'Minimalist Bayesian Framework' for stochastic optimization. This framework simplifies traditional Bayesian methods by placing a prior only...
Research & Development
Navigating AI Maturity: A New Framework for Responsible Engagement
TLDR: The Hierarchy of Engagement with AI (HE-AI) model, inspired by Maslow's hierarchy of needs, introduces an eight-level framework for individuals and organizations to...