Lipreading, also known as visual speech recognition, aims to identify the speech content from videos by analyzing the visual deformations of lips and nearby areas. One of the significant obstacles for research in this field is the lack of proper datasets for a wide variety of languages: so far, these methods have been focused only on English or Chinese. In this paper, we introduce a naturally distributed large-scale benchmark for lipreading in Russian language, named LRWR, which contains 235 classes and 135 speakers. We provide a detailed description of the dataset collection pipeline and dataset statistics. We also present a comprehensive comparison of the current popular lipreading methods on LRWR and conduct a detailed analysis of their performance. The results demonstrate the differences between the benchmarked languages and provide several promising directions for lipreading models finetuning. Thanks to our findings, we also achieved new state-of-the-art results on the LRW benchmark.
We present Catalyst.RL, an open-source PyTorch framework for reproducible and sample efficient reinforcement learning (RL) research. Main features of Catalyst.RL include large-scale asynchronous distributed training, efficient implementations of various RL algorithms and auxiliary tricks, such as n-step returns, value distributions, hyperbolic reinforcement learning, etc. To demonstrate the effectiveness of Catalyst.RL, we applied it to a physics-based reinforcement learning challenge “NeurIPS 2019: Learn to Move – Walk Around” with the objective to build a locomotion controller for a human musculoskeletal model. The environment is computationally expensive, has a high-dimensional continuous action space and is stochastic. Our team took the 2nd place, capitalizing on the ability of Catalyst.RL to train high-quality and sample-efficient RL agents in only a few hours of training time. The implementation along with experiments is open-sourced so results can be reproduced and novel ideas tried out.
In the NeurIPS 2018 Artificial Intelligence for Prosthetics challenge, participants were tasked with building a controller for a musculoskeletal model with a goal of matching a given time-varying velocity vector. Top participants described their algorithms in this paper. Many solutions use similar relaxations and heuristics, such as reward shaping, frame skipping, discretization of the action space, symmetry, and policy blending. However, each team implemented different modifications of the known algorithms by, for example, dividing the task into subtasks, learning low-level control, or by incorporating expert knowledge and using imitation learning.
Despite the recent progress in deep reinforcement learning field (RL), and, arguably because of it, a large body of work remains to be done in reproducing and carefully comparing different RL algorithms. We present catalyst. RL, an open source framework for RL research with a focus on reproducibility and flexibility. Main features of our library include large-scale asynchronous distributed training, easy-to-use configuration files with the complete list of hyperparameters for the particular experiments, efficient implementations of various RL algorithms and auxiliary tricks, such as frame stacking, n-step returns, value distributions, etc. To vindicate the usefulness of our framework, we evaluate it on a range of benchmarks in a continuous control, as well as on the task of developing a controller to enable a physiologically-based human model with a prosthetic leg to walk and run. The latter task was introduced at NeurIPS 2018 AI for Prosthetics Challenge, where our team took the 3rd place, capitalizing on the ability of catalyst. RL to train high-quality and sample-efficient RL agents.
In the NIPS 2017 Learning to Run challenge, participants were tasked with building a controller for a musculoskeletal model to make it run as fast as possible through an obstacle course. Top participants were invited to describe their algorithms. In this work, we present eight solutions that used deep reinforcement learning approaches, based on algorithms such as Deep Deterministic Policy Gradient, Proximal Policy Optimization, and Trust Region Policy Optimization. Many solutions use similar relaxations and heuristics, such as reward shaping, frame skipping, discretization of the action space, symmetry, and policy blending. However, each of the eight teams implemented different modifications of the known algorithms.
In this paper, we present our approach to solve a physics-based reinforcement learning challenge “Learning to Run'' with objective to train physiologically-based human model to navigate a complex obstacle course as quickly as possible.The environment is computationally expensive, has a high-dimensional continuous action space and is stochastic. We benchmark state of the art policy-gradient methods and test several improvements, such as layer normalization, parameter noise, action and state reflecting, to stabilize training and improve its sample-efficiency.We found that the Deep Deterministic Policy Gradient method is the most efficient method for this environment and the improvements we have introduced help to stabilize training.Learned models are able to generalize to new physical scenarios, e.g. different obstacle courses.