About
Research & Development
Advancing Federated Graph Learning with Semi-Asynchronous Communication and Intelligent Client Collaboration
TLDR: FedSA-GCL is a novel semi-asynchronous framework for federated graph learning that addresses the inefficiencies of synchronous methods and the limitations of existing asynchronous...
Research & Development
Bridging the Gap: How Reinforcement Learning Can Accurately Model Economic Markets
TLDR: This paper addresses two key biases—structural and parametric—that arise when applying Reinforcement Learning (RL) to economic models. The structural bias causes RL agents...
Research & Development
Unveiling DepthDark: Accurate Depth Mapping for Nighttime Scenes
TLDR: DepthDark is a novel foundation model for monocular depth estimation specifically designed for low-light environments. It addresses the lack of suitable training data...
Research & Development
Improving Bangla Punctuation with AI: A New Approach for Low-Resource Languages
TLDR: This research explores using transformer models, specifically XLM-RoBERTa-large, to automatically restore punctuation in unpunctuated Bangla text. Addressing the scarcity of annotated data, the...
Research & Development
Bridging the Gap: Enhancing Digital Twin Performance with Bidirectional Knowledge Transfer
TLDR: This research introduces a new module for Digital Twin frameworks that quantifies and manages the "reality gap" between simulations and real-world data. By...
Research & Development
C2G-KD: Generating Synthetic Data with Geometric Constraints for AI Training
TLDR: C2G-KD is a novel data-free knowledge distillation framework that trains a class-conditional generator to produce synthetic samples. It uses PCA-derived geometric constraints, estimated...