Perceivable-Human-Robot-Interaction-using-Neural-Attention-Q-Networks
Robot teacher
An implementation of a deep reinforcement learning algorithm to teach a robot to interact with humans in a socially acceptable manner.
Multimodal Deep Attention Recurrent Q-Network for perceivable social human-robot interaction.
4 stars
2 watching
0 forks
Language: Lua
last commit: over 7 years ago activity-recognitionactivity-understandingattention-mechanismdeep-reinforcement-learninghuman-robot-interactionperceptionreinforcement-learningsocial-interaction-skills
Related projects:
| Repository | Description | Stars |
|---|---|---|
| ahq1993/multimodal-deep-q-network-for-social-human-robot-interaction | A deep reinforcement learning system designed to enable robots to acquire social interaction skills through multimodal interactions with humans. | 14 |
 5vision/darqn | An implementation of a deep reinforcement learning model for continuous control tasks | 116 |
 vita-epfl/crowdnav | Develops robot navigation policies in crowded spaces using reinforcement learning and attention mechanisms. | 607 |
 jhu-lcsr/good_robot | Research on repurposing reinforcement learning for transfer between tasks in robotics and multi-step visual tasks with simulation-to-real transfer | 108 |
 makrisio/q-learning-algorithm-implementation-in-matlab | An implementation of the Q-Learning algorithm in MATLAB for training agents to navigate mazes | 43 |
 minqi/learning-to-communicate-pytorch | This project implements a PyTorch-based framework for learning discrete communication protocols in multi-agent reinforcement learning environments. | 349 |
 kuz/deepmind-atari-deep-q-learner | An implementation of a deep reinforcement learning architecture for playing Atari games | 1,828 |
 hungtuchen/pytorch-dqn | An implementation of a deep reinforcement learning network using PyTorch to learn human-level control through trial and error. | 387 |
 google-deepmind/meltingpot | Assesses generalization of multi-agent reinforcement learning algorithms to novel social situations | 637 |
 robot-perception-group/airpose | A deep learning framework for multi-view fusion network-based 3D human motion capture system | 52 |
 intelligent-machine-learning/dlrover | Automates large-scale deep learning training on distributed clusters, providing fault tolerance and fast recovery from failures. | 1,302 |
 instance01/qubo-nn | A tool for training neural networks on Quantum Annealing (QA) optimization problems represented as QUBO matrices. | 47 |
 berkeleyautomation/gqcnn | Develops and analyzes convolutional neural networks for grasp quality in robotics | 316 |
 thealexhong/starship | An architecture for affective human-robot interaction using multimodal recognition and emotional response models. | 2 |
 aiqm/torchani | A PyTorch implementation of a neural network potential for molecular simulations | 471 |