TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker and becoming a Supporting Member. For more info: Support TMC
Start a Discussionhttps://teslamotorsclub.com/tmc/tags/

Harvard / MIT machine learning research

Discussion in 'Electric Vehicles' started by adamjb, Oct 3, 2017.

  1. adamjb

    adamjb Member

    Joined:
    Oct 3, 2017
    Messages:
    5
    Location:
    Cambridge
    New to the forum - wanted to reach out to Tesla owners that might be interested in participating in research that I'm working on with a group of students at MIT and Harvard.

    We're working with Thomas Poggio and Vikash Mansinghka on a project to apply Machine Learning / AI techniques to novel datasets and applications (see: 9.520/6.860, Fall 2017), with an interest in driving data.

    Our goal is to work with driving datasets from individual drivers (via access to Tesla API keys), in addition to aggregated datasets from Uber and Lyft. We're hoping to help people improve their driving, optimize charging & routes, and discover cool & useful insights.

    If interested, we made a quick online form to sign up: carriage.

    Feel free to reply with questions, or email me directly at adamjb at mit dot edu.

    Thanks in Advance!

    Adam, Jifei, Mike, and Alex (our team that's working on this, whose bios you can read in more depth via the above link)
     
  2. BizJet

    BizJet Member

    Joined:
    May 19, 2016
    Messages:
    108
    Location:
    Boston
    All --

    I know the leader of the team who's conducting this study. They're solid, reliable graduate students at Harvard and MIT who are committed to really using machine learning to help give insights to us Tesla drivers. I'm participating, and I recommend this to my fellow Tesla drivers.

    Michael
     
  3. adamjb

    adamjb Member

    Joined:
    Oct 3, 2017
    Messages:
    5
    Location:
    Cambridge
    Thanks, Michael! And the link in the above is broken -- if folks are interested they can sign up at teslamlresearch.com
     
  4. adamjb

    adamjb Member

    Joined:
    Oct 3, 2017
    Messages:
    5
    Location:
    Cambridge
    And just for reference for people, we also came up with a short protocol to communicate to participants (with help from the MIT Age Lab who has done research with Tesla Drivers):
    1. You own all of your data. Which means you can request it’s deletion or transfer at any time.
    2. We will not publish any raw data or make public. If we decide to publish our algorithms in a paper, we will request approval from all participants ahead of time.
    3. We will only access features of the API that directly relate to vehicle perform and location—gps, charge data, efficiency, etc. We will not access any other features such as the internal environment variables.
    4. We plan to access your API for a period of 2 months in order to gather a sufficient data set. After this period we will delete your API key from our system and notify you of its deletion.
    5. At any point, you can request deletion of your key and removal from the research study.
    6. If you happen to have an accident or any sort of violation, we will not submit any data to authorities and will delete all data on file that relate to your vehicle upon request.
    As always, also happy to answer questions people have directly via email at adamjb at mit dot edu, or via the forum.
     

Share This Page