Skip to main content
Intel Jobs Home
  • Discover Intel

    • Life at Intel
    • Innovative Technology
    • Diversity
    • US Veterans
    • Education Visionaries
    • Corporate Responsibilities
    • Intel Around the World
    • Benefits at Intel

    Advice & Help

    • Hiring Process & Tips
    • Candidate Help Desk
    • Interviewing at Intel

    Global Student Center

    • Explore Global Student Center
    • Internships
    • Rotation Programs

    Other Languages

    • Hebrew Career-site

    Explore

    • Explore more careers
  • Jobs Areas

    Software

    • Software Application
    • Firmware and BIOS
    • Network Engineering
    • Operating System Engineer
    • Systems Engineering
    • Test, Validation, and Verification Engineering
    • Linux Kernel Engineering
    • RF Design
    • Software Development Engineering
    • Software Internships

    Manufacturing and Facilities

    • Manufacturing Process Engineering
    • Manufacturing Technician
    • Corporate Services
    • Supply Chain & Materials
    • Module Integration & Yield Engineer
    • Quality and Reliability Engineering
    • Equipment Technician
    • Equipment Engineering
    • Manufacturing and Facilities Internships

    Hardware

    • Hardware Logic Design
    • Analog Engineering
    • Physical Design Engineering
    • FPGA
    • Quality and Reliability Engineering
    • Hardware Research and Development
    • Platform Engineering
    • System Validation
    • Hardware Internships

    SoC Design

    • SoC Design Engineering
    • RTL Integration
    • SoC Physical Design
    • SoC Verification Engineering
    • Digital Design
    • Memory Design and Development
    • Structural Design
    • SoC Design Internships

    Business

    • Org Design
    • Talent Acquisition
    • HR Specialist
    • Project and Program Managing
    • Operations
    • Financial Analyst
    • Accounting
    • Business Internships

    Artificial Intelligence

    • Data Science
    • AI Software
    • AI Hardware
    • AI Business
    • AI Research
    • AI Internships
    • AI Residency

    Graphics

    • Graphics Research
    • Graphics Architecture
    • Graphics Software
    • Graphics Validation and Verification
    • Graphics Hardware
    • Graphics Internships

    Silicon Photonics

    • RX Chip Design and Integration
    • Photonic Integrated Circuit Design Engineering
    • Optical Engineering
    • Laser Engineering
    • Optical Testing
    • Silicon Photonics Internships

    Information Technology

    • IT Security
    • Data Analyst
    • Support Technician
    • Network Specialist
    • IT Internships

    Sales and Marketing

    • Customer and Partner Sales
    • Sales Support
    • Technical Sales
    • Marketing
    • Marketing Internships

    Connectivity

    • Modem Design
    • Modem Verification
    • Software Protocol Engineering
    • Physical Design
    • RTL Integration and Micro Architecture Engineering
    • Connectivity Internships
close
 

Announcing Reinforcement Learning Coach v1.0.0 – Batch RL, More Algos, New APIs

Hear more

Gal Leibovich
Gal Novik
Shadi Endrawis

Ever since we built and released RL Coach, our open source framework for training and evaluating reinforcement learning agents, in 2017, we have been working hard to add algorithms, simulation environments and features that will make it useful for the machine learning research and engineering communities. Features such as benchmarks, native support for hierarchical RL, and horizontal scaling helped us demonstrate a strong and extensible foundation for agent development and training. During 2018 we also integrated Coach with AWS Sagemaker, where it is used to train DeepRacer and solve other challenges by AWS and their customers. We’re very happy to see the growing usage of Coach by both researchers [1], [2], [3] and engineers [4], [5] to design new algorithms or build RL-based solutions, and would love to hear how Coach helped you in your project at coach@intel.com.

In the past few months we have taken additional steps to bring RL to more use cases than research and to grow the community of Coach users. The latest additions to Coach go beyond simulation-based learning environments, incorporate newer and stronger RL algorithms, and maintain and extend the APIs to improve usability. Today, we are very excited to announce the 1.0.0 release of RL Coach. The new release features the implementation of several new algorithms (for a total of 27), support for Batch Reinforcement Learning, improved documentation, bug fixes and new APIs that enable the use of Coach as a Python library. With the 1.0.0 release we believe that the main software structure of Coach has matured and stabilized and no major API changes are on the horizon.

Batch Reinforcement Learning

Many real-world problems are missing a simulator to accurately model the environment that the agent would interact with in a standard reinforcement learning setting. Often, all a data scientist may have is data that was collected using a deployed policy, and this existing data must be used to learn a better policy for solving the problem. One such example is improving drug dose management or drug admission scheduling policy for patients. In these situations, we have data based on the policy that was used with previous patients, but we cannot conduct additional experiment on the same patients to collect new data. Here is where batch reinforcement learning allows RL to learn from a dataset, while also exercising the dataset for off-policy evaluation of the goodness of the learned policy.

With the 1.0.0 release, we have added support for batch reinforcement learning in Coach, while also enabling off-policy evaluation (OPE) of the learned policy based on data that was acquired using another policy. We have added several off-policy evaluators for contextual bandits (Direct Method, Inverse Propensity Scoring and Doubly Robust) and for reinforcement learning (Sequential Doubly Robust and Weighted Importance Sampling), while allowing the use of a wide range of integrated off-policy RL algorithms in Coach. We also added support for a variant of the Batch Constrained Q Learning algorithm for discrete action space problems. We encourage you to try it out with our deep dive tutorial on Batch Reinforcement Learning.

New Algorithms

Since our last blog post, we have added support for several new reinforcement learning agents: Sample Efficient Actor-Critic with Experience Replay (ACER), Soft Actor-Critic (SAC) and Twin-Delayed Deep Deterministic Policy Gradient (TD3). As always, when we add new RL algorithms to Coach, we aim to fully reproduce paper results, as shown in Coach Benchmarks. This is also the case with the newly added algorithms.

We’d be happy to get feedback on additional features that may be useful and on your experience using Coach. You can contact us at coach@intel.com or on our GitHub repo. We’d also appreciate any contributions that can be useful for other members of the machine learning community.

Ready to check out the new Coach release? You can get started by cloning the repository and running through our Getting Started tutorial. For the latest advancements from the Intel AI research team, visit intel.ai/research and follow us on Twitter: @IntelAIResearch.

Yang, Process Equipment Engineer

One of the promising reinforcement learning research areas is Hierarchical Reinforcement Learning (HRL). Papers such as "Hierarchical Actor-Critic" and "Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation" use agents operating in different time scales or goal complexity to achieve the desired high-level goals.
We have now added multi-agent support to Reinforcement Learning Coach, allowing the invocation of several agents training together. This is first being exemplified by the addition of generic HRL support to Reinforcement Learning Coach and specifically with the implementation of the Hierarchical Actor-Critic paper.

[1] https://arxiv.org/pdf/1904.01790.pdf
[2] https://www.sciencedirect.com/science/article/pii/S0925231219308367
[3] http://openaccess.thecvf.com/content_ECCV_2018/papers/Xiaodan_Liang_CIRL_Controllable_Imitative_ECCV_2018_paper.pdf
[4] https://aws.amazon.com/blogs/machine-learning/an-introduction-to-reinforcement-learning-with-aws-robomaker/
[5] https://www.analyticsindiamag.com/researchers-are-now-giving-the-sense-of-touch-to-robots-through-deep-learning/

Original article from:
https://www.intel.com/content/www/us/en/artificial-intelligence/posts/rl-coach-new-release.html?wapkw=coach

Hear more about our technologies, sign up here

Back to homepage

Share this job on social media

  • Software Careers
  • Software Application
  • Firmware and BIOS
  • Network Engineering
  • Operating System Engineer
  • Systems Engineering
  • Test, Validation, and Verification Engineering
  • Linux Kernel Engineering
  • RF Design
  • Software Development Engineering
  • Software Internships
  • Artificial Intelligence
  • Data Science
  • AI Software
  • AI Hardware
  • AI Business
  • AI Research
  • AI Internships and Residency
  • Hardware
  • Hardware Logic Design
  • Analog Engineering
  • Physical Design Engineering
  • FPGA
  • Quality and Reliability Engineering
  • Hardware Research and Development
  • Platform Engineering
  • System Validation
  • Hardware Internships
  • Information Technology
  • IT Security
  • Data Analyst
  • Support Technician
  • Network Specialist
  • IT Internships
  • Manufacturing and Facilities
  • Manufacturing Process Engineering
  • Manufacturing Technician
  • Corporate Services
  • Supply Chain & Materials
  • Module Integration & Yield Engineer
  • Quality and Reliability Engineering
  • Assembly & Test Manufacturing Engineer
  • Equipment Technician
  • Equipment Engineering
  • Manufacturing and Facilities Internships
  • SoC Design
  • SoC Design Engineering
  • RTL Integration
  • SoC Physical Design
  • SoC Verification Engineering
  • Digital Design
  • Memory Design and Development
  • Structural Design
  • SoC Design Internships
  • Sales and Marketing
  • Customer and Partner Sales
  • Sales Support
  • Technical Sales
  • Marketing
  • Marketing Internships
  • Business
  • Org Design
  • Talent Acquisition
  • HR Specialist
  • Project and Program Managing
  • Operations
  • Financial Analyst
  • Accounting
  • Business Internships
  • Graphics
  • Graphics Research
  • Graphics Architecture
  • Graphics Software
  • Graphics Validation and Verification
  • Graphics Hardware
  • Graphics Internships
  • Silicon Photonics
  • RX Chip Design and Integration
  • Photonic Integrated Circuit Design Engineering
  • Optical Engineering
  • Laser Engineering
  • Optical Testing
  • Silicon Photonics Internships
  • Connectivity
  • Modem Design
  • Modem Verification
  • Software Protocol Engineering
  • Physical Design
  • RTL Integration and Micro Architecture Engineering
  • Connectivity Internships
  • Go back to Home

Grow your network of opportunities

Facebook Twitter LinkedIn Glassdoor Instagram
Intel
  • Company Information
  • Our Commitment
  • Investor Relations
  • Contact Us
  • Newsroom
  • Site Map
  • Jobs
  • © Intel Corporation
  • Terms of Use
  • *Trademarks
  • Privacy
  • Cookies
  • Supply Chain Transparency
  • Powered by SmashFly
Intel
  1. Home
  2. Career

Recommended Jobs