Charles (Chip) Schaff

Toyota Technological Institute at Chicago (TTIC)

6045 S. Kenwood Ave., Chicago, IL 60637

cbschaff (at) ttic (dot) edu

Resume  •  Google Scholar

I am on the job market looking for research or engineering positions related to Reinforcement Learning, Deep Learning, and Robotics.

I am a final-year PhD student at Toyota Technological Institute at Chicago (TTIC) studying Machine Learning and Robotics under the supervision of Matthew Walter. Before coming to TTIC, I recieved my B.S. in Computer Science from Washington University in St Louis in 2016.

Publications

• Soft Robots Learn to Crawl: Jointly Optimizing Design and Control with Sim-to-Real Transfer
  Charles Schaff, Audrey Sedal, Matthew R. Walter
  Robotics: Science and Systems (RSS), 2022
  [arxiv] [code] [project page]

• Benchmarking Structured Policies and Policy Optimization for Real-World Dexterous Object Manipulation
  Niklas Funk, Charles Schaff, Rishabh Madan, Takuma Yoneda, Julen Urain De Jesus, Joe Watson, Ethan K. Gordon,
  Felix Widmaier, Stefan Bauer, Siddhartha S. Srinivasa, Tapomayukh Bhattacharjee, Matthew R. Walter, Jan Peters
  IEEE Robotics and Automation Letters (RA-L), 2021
  [arxiv] [code] [project page]

• Residual Policy Learning for Shared Autonomy
  Charles Schaff and Matthew R. Walter
  Robotics: Science and Systems (RSS), 2020
  [arxiv] [code] [project page]

• Jointly Learning to Construct and Control Agents using Deep Reinforcement Learning
  Charles Schaff, David Yunis, Ayan Chakrabarti, and Matthew R. Walter
  IEEE International Conference on Robotics and Automation (ICRA), 2019
  [arxiv] [code] [project page]

• Jointly Optimizing Placement and Inference for Beacon-based Localization
  Charles Schaff, David Yunis, Ayan Chakrabarti, and Matthew R. Walter
  IEEE International Conference on Intelligent Robots and Systems (IROS), 2017
  [arxiv] [code] [project page]

• Bayesian Optimization for Automated Model Selection
  Gustavo Malkomes, Charles Schaff, Roman Garnett
  Conference on Neural Information Processing Systems (NuerIPS), 2016
  [NeurIPS]

Projects and Events

• Fall 2020 - Dexterous Manipulation with the TriFinger Platform
  
  

  We participated in the Real Robot Challenge which tasked participants with manipulating objects with the TriFinger Platform. We won 1st place in all competitive phases and recieved a 12,500 Euro cash prize.

  [code] [paper] [videos]
• April 2019 - Baxter solving a Rubik's Cube, 2019 National Robotics Week at MSI
  
  

  I led other members of the RIPL lab at the TTIC in the development of software to physically solve a Rubik's Cube with our Baxter robot. We showcased our system at the Museum of Science and Industry in Chicago for the 2019 National Robotics Week exhibit. Using only cameras mounted on Baxer our robot was able to find cubes on a table, pick them up, scan the cube to determine its state, and manipulate the cube until solved. We used clustering techniques to determine the colors on each face of the cube, Kociemba's two phase algorithm for solving the cube, and visual servoing to accurately turn cube faces.

  [videos]

---

• April 2018 - UR5-equipped Robot playing Checkers, 2018 National Robotics Week at MSI
  
  

  With Prof. Matthew Walter and other students in the RIPL lab at the TTIC, I developed and showcased our UR5-equipped Husky A200 robot safely playing Checkers against human opponents at the Museum of Science and Industry in Chicago for the 2018 National Robotics Week exhibit. Using an Realsense RGBD camera, the robot determined the location of the pieces, the state of the game, and presense of of humans in its workspace. The robot played the game using a simple minimax agent and was able to set clear and set up the pieces.

---

• November 2017 - Transfer Learning for Autonomous Driving in Duckietown
  
  

  The Duckietown project aims to educate and excite the world about the challenges of robotics and AI by building and experimenting with autonomous vehicles called "Duckiebots" in a miniature town. In the fall of 2017, I participated in a Duckietown course taught by Prof. Matthew Walter at TTIC. My project in that course centered around transferring models for autonomous driving trained in simulation to the real Duckietown. For more details and videos, check out the project page.

  [project page]

Work Experience

• Intern at Google Inc. in Sunnyvale, CA (Summer 2019)
  I worked with several teams across Google and DeepMind to explore how Reinforcement Learning could be used to improve video compression.

• Intern at Google Inc. in Mountain View, CA (Summer 2015)
  I worked with the Google Maps team to develop Machine Learning techniques for Google's semantic location ranking system. Specifically, I used decision trees to improve the capability of their feature engineering pipeline.

• Intern at Epic Systems in Verona, WI (Summer 2014)
  I worked with Epic's transplant team, Pheonix, to develop algorithms for efficient kidney allocation in paired kidney exchanges. I worked with another intern to implement different graph based algorithms and integrate our solution into the Epic system.

Graduate Coursework

• 2018 - TTIC 31070: Convex Optimization
• 2017 - TTIC 31240: Self-driving Vehicles: Models and Algorithms for Autonomy
• 2017 - TTIC 31170: Planning, Learning and Estimation for Robotics, AI
• 2017 - TTIC 31220: Unsupervised Learning and Data Analysis
• 2017 - TTIC 31230: Fundamentals of Deep Learning
• 2017 - TTIC 31010: Algorithms
• 2016 - TTIC 31120: Statistical and Computational Learning Theory
• 2016 - TTIC 31020: Introduction to Statistical Machine Learning
• 2016 - TTIC 31150: Mathematical Toolkit
• 2015 - WUSTL CSE 515T: Bayesian Methods in Machine Learning
• 2014 - WUSTL CSE 519T: Advanced Machine Learning
• 2014 - WUSTL CSE 517A: Machine Learning
• 2013 - WUSTL CSE 511A: Introduction to Artificial Intelligence

Teaching Experience

• 2019 - TTIC 31170: Planning, Learning and Estimation for Robotics and AI, Teaching Assistant
• 2017 - TTIC 31150: Mathematical Toolkit, Teaching Assistant
• 2015 - WUSTL CSE 517A: Machine Learning, Teaching Assistant
• 2015 - WUSTL CSE 511A: Introduction to Artificial Intelligence, Teaching Assistant
• 2014 - WUSTL CSE 511A: Introduction to Artificial Intelligence, Teaching Assistant
• 2014 - WUSTL CSE 417T: Introduction to Machine Learning, Teaching Assistant
• 2013 - WUSTL CSE 260: Introduction to Digital Logic and Computer Design, Teaching Assistant

Last Updated: April, 2022