Abstract: We consider the use of polar codes in edge computing and communications applications, and to achieve energy efficiency, we propose implementation of neural network-based decoders for them ...

Abstract: This paper features a power-efficient Viterbi decoder targeted for energy-efficient digital communication systems such as mobile and edge applications. An attempt has been made at a ...

A Flutter mobile application that performs interactive, on-device image segmentation using the EdgeTAM model. Users can select an image from their gallery, tap on an object to create a segmentation ...

Protiformer is a deep learning framework for natural language-guided protein engineering. It combines contrastive learning (CLAP) and graph neural networks to enable protein design based on text ...