Tushar K

Tushar K

San Francisco, California, United States
1K followers 500+ connections

About

Currently working as a Software Engineer at Nimble Robotics. Graduated with my Master's…

Activity

Join now to see all activity

Experience

  • Nimble Robotics Graphic

    Nimble Robotics

    San Francisco Bay Area

  • -

    Los Angeles

  • -

    Remote

  • -

    Los Angeles, California, United States

  • -

    Los Angeles, California, United States

  • -

    Bengaluru Area, India

  • -

    Bengaluru, Karnataka, India

  • -

    Bengaluru Area, India

Education

  • University of Southern California Graphic

    University of Southern California

    -

    Activities and Societies: Swimming, Autonomous Vehicle Club, Financial Investing Club

    a) Part of the Robotic Embedded Systems Lab (RESL) and Autonomous Vehicle Lab (AVL).
    b) Part of the University Advancement working as a Data Analyst.
    c) Courses focused on Machine Learning and its applications with a concentration in Reinforcement Learning.

  • -

    Activities and Societies: Quiz Club, Swimming, Waterpolo

    1)Headed the Quiz club
    2)Represented University at the University Nationals in Waterpolo
    3)Runner up at TCS Tech Bytes

Publications

  • Sample Factory: Egocentric 3D Control from Pixels at 100000 FPS with Asynchronous Reinforcement Learning

    International Conference on Machine Learning (ICML) 2020

    Increasing the scale of reinforcement learning experiments has allowed researchers to achieve unprecedented results in both training sophisticated agents for video games, and in sim-to-real transfer for robotics. Typically such experiments rely on large distributed systems and require expensive hardware setups, limiting wider access to this exciting area of research. In this work, we aim to solve this problem by optimizing the efficiency and resource utilization of reinforcement learning…

    Increasing the scale of reinforcement learning experiments has allowed researchers to achieve unprecedented results in both training sophisticated agents for video games, and in sim-to-real transfer for robotics. Typically such experiments rely on large distributed systems and require expensive hardware setups, limiting wider access to this exciting area of research. In this work, we aim to solve this problem by optimizing the efficiency and resource utilization of reinforcement learning algorithms instead of relying on distributed computation. We present the "Sample Factory", a high-throughput training system optimized for a single-machine setting. Our architecture combines a highly efficient, asynchronous, GPU-based sampler with off-policy correction techniques, allowing us to achieve throughput higher than 105 environment frames/second on non-trivial control problems in 3D without sacrificing sample efficiency. We extend Sample Factory to support self-play and population-based training and apply these techniques to train highly capable agents for a multiplayer first-person shooter game.

    Other authors
    See publication

Projects

  • Faster fifo

    - Present

    A faster alternative to Python's standard multiprocessing.Queue (IPC FIFO queue). Up to 30x faster in some configurations. Available to download as a pip package.

    Other creators
    See project
  • Sample Factory

    - Present

    A package for high throughput asynchronous Reinforcement Learning

    Other creators
    See project
  • TravBud

    -

    A Social Network for Travellers.

    It combines the power of Social Networking and the machine learning algorithms to deliver an elegant experience to the users of this website

    Tools used:
    Google App Engine
    Amazon Web Services

    Programming Languages:
    Python
    HTML5, CSS3, JS

    I was responsible for the backend development of this application with Python.

    Other creators

More activity by Tushar

View Tushar’s full profile

  • See who you know in common
  • Get introduced
  • Contact Tushar directly
Join to view full profile

Other similar profiles

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Others named Tushar K in United States

Add new skills with these courses