Turing NLG
Turing NLG is a large-scale natural language generation model developed by Microsoft. It is designed to generate human-like text and perform various language understanding tasks with high accuracy.
deep learning
MetricGAN+ (speech enhancement metric learning)
MetricGAN+ is an advanced framework for speech enhancement that leverages metric learning and generative adversarial networks to optimize speech quality metrics directly. It improves upon its predecessor, MetricGAN, by enhancing performance and stability in denoising and speech enhancement tasks.
Neural animation
Neural animation refers to the use of neural networks and machine learning techniques to generate, enhance, or control animated content. This technology applies artificial intelligence to automate or assist in the creation of animations, enabling new possibilities in fields such as film, gaming, and virtual reality.
Caffe
Caffe is a deep learning framework developed for machine learning and computer vision tasks. It is known for its speed, modularity, and expressive architecture, widely used in academic research and industry applications.
DEMUCS (denoising music source separation)
DEMUCS is a deep learning model designed for music source separation with a focus on denoising, enabling the extraction of individual instrument and vocal tracks from mixed audio recordings. It uses convolutional and recurrent neural network architectures to achieve high-quality separation and noise reduction.
Neural network pruning
Neural network pruning is a technique in machine learning used to reduce the size and complexity of artificial neural networks by removing unnecessary parameters or connections. This process aims to improve efficiency, decrease computational requirements, and sometimes enhance generalization without significantly compromising accuracy.
Theano (software)
Theano is an open-source Python library designed for efficient numerical computation, particularly in large-scale machine learning tasks. It allows users to define, optimize, and evaluate mathematical expressions involving multi-dimensional arrays.
GET3D (generative 3D model)
GET3D is a generative 3D model designed to produce high-quality three-dimensional objects from 2D images. It leverages advances in machine learning to generate detailed 3D shapes and textures, enabling applications in graphics, gaming, and virtual reality.
Expectation–maximization for deep learning
Expectation–maximization for deep learning is a computational approach that integrates the EM algorithm into neural network training to handle incomplete data and latent variable models. It iteratively estimates missing data and optimizes model parameters, improving learning in unsupervised and semi-supervised contexts.
AlphaFold
AlphaFold is an artificial intelligence system developed to predict protein structures from amino acid sequences with high accuracy. It represents a significant advancement in computational biology, enabling faster and more reliable protein modeling than previous methods.