Reinforcement Learning

Jul 08, 2022

Designing Arithmetic Circuits with Deep Reinforcement Learning

As Moore’s law slows down, it becomes increasingly important to develop other techniques that improve the performance of a chip at the same technology process...
9 MIN READ
A large compilation of Minecraft videos that MineDojo uses to train the AI
Jul 01, 2022

Building Generally Capable AI Agents with MineDojo

Using video games as a medium for training AI has become a popular method within the AI research community. These autonomous agents have had great success in...
6 MIN READ
Jun 30, 2022

The Full Stack Optimization Powering NVIDIA MLPerf Training v2.0 Performance

MLPerf benchmarks are developed by a consortium of AI leaders across industry, academia, and research labs, with the aim of providing standardized, fair, and...
14 MIN READ
Sep 22, 2021

NVIDIA Research: Transferring Dexterous Manipulation from GPU Simulation to a Remote, Real-World, TriFinger Task

A critical question to ask when designing a machine learning–based solution is, “What’s the resource cost of developing this solution?” There are...
12 MIN READ
Nov 05, 2020

Discovering GPU-friendly Deep Neural Networks with Unified Neural Architecture Search

After the first successes of deep learning, designing neural network architectures with desirable performance criteria for a given task (for example, high...
9 MIN READ
Sep 01, 2020

Enhancing Sample Efficiency in Reinforcement Learning with Nonparametric Methods

Recent developments in artificial intelligence and autonomous learning have shown impressive results in tasks like board games and computer games. However, the...
10 MIN READ
Apr 15, 2020

Powering AutoML-enabled AI Model Training with Clara Train

Deep neural networks (DNNs) have been successfully applied to volume segmentation and other medical imaging tasks. They are capable of achieving...
8 MIN READ
Figure 1: Value iteration constructs the value function over all states over time. Here each square is a state: S is the start state, G the goal state, T squares are traps, and black squares cannot be entered. In value iteration we initialize the rewards (traps and goal state) and then these reward values spread over time until an equilibrium is reached. Depending on the penalty value on traps and the reward value for the goal different solution patterns might emerge; the last two grids show such solution states.
Sep 08, 2016

Deep Learning in a Nutshell: Reinforcement Learning

This post is Part 4 of the Deep Learning in a Nutshell series, in which I’ll dive into reinforcement learning, a type of machine learning in which agents take...
28 MIN READ
Apr 27, 2016

Train Your Reinforcement Learning Agents at the OpenAI Gym

Today OpenAI, a non-profit artificial intelligence research company, launched OpenAI Gym, a toolkit for developing and comparing reinforcement...
9 MIN READ