Close Menu
    Trending
    • Why Should We Bother with Quantum Computing in ML?
    • Federated Learning and Custom Aggregation Schemes
    • How To Choose The Perfect AI Tool In 2025 » Ofemwire
    • Implementing DRIFT Search with Neo4j and LlamaIndex
    • Agentic AI in Finance: Opportunities and Challenges for Indonesia
    • Dispatch: Partying at one of Africa’s largest AI gatherings
    • Topp 10 AI-filmer genom tiderna
    • OpenAIs nya webbläsare ChatGPT Atlas
    ProfitlyAI
    • Home
    • Latest News
    • AI Technology
    • Latest AI Innovations
    • AI Tools & Technologies
    • Artificial Intelligence
    ProfitlyAI
    Home » New tool evaluates progress in reinforcement learning | MIT News
    Artificial Intelligence

    New tool evaluates progress in reinforcement learning | MIT News

    ProfitlyAIBy ProfitlyAIMay 6, 2025No Comments6 Mins Read
    Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
    Share
    Facebook Twitter LinkedIn Pinterest Email

    If there’s one factor that characterizes driving in any main metropolis, it’s the fixed stop-and-go as visitors lights change and as automobiles and vans merge and separate and switch and park. This fixed stopping and beginning is extraordinarily inefficient, driving up the quantity of air pollution, together with greenhouse gases, that will get emitted per mile of driving. 

    One method to counter this is called eco-driving, which may be put in as a management system in autonomous autos to enhance their effectivity.

    How a lot of a distinction might that make? Would the affect of such techniques in decreasing emissions be well worth the funding within the know-how? Addressing such questions is one in all a broad class of optimization issues which have been troublesome for researchers to handle, and it has been troublesome to check the options they provide you with. These are issues that contain many alternative brokers, resembling the numerous completely different sorts of autos in a metropolis, and various factors that affect their emissions, together with velocity, climate, highway situations, and visitors gentle timing.

    “We acquired a couple of years in the past within the query: Is there one thing that automated autos might do right here when it comes to mitigating emissions?” says Cathy Wu, the Thomas D. and Virginia W. Cabot Profession Growth Affiliate Professor within the Division of Civil and Environmental Engineering and the Institute for Information, Techniques, and Society (IDSS) at MIT, and a principal investigator within the Laboratory for Data and Resolution Techniques. “Is it a drop within the bucket, or is it one thing to consider?,” she questioned.

    To deal with such a query involving so many parts, the primary requirement is to collect all out there information concerning the system, from many sources. One is the format of the community’s topology, Wu says, on this case a map of all of the intersections in every metropolis. Then there are U.S. Geological Survey information exhibiting the elevations, to find out the grade of the roads. There are additionally information on temperature and humidity, information on the combo of auto sorts and ages, and on the combo of gas sorts.

    Eco-driving includes making small changes to reduce pointless gas consumption. For instance, as automobiles method a visitors gentle that has turned crimson, “there’s no level in me driving as quick as doable to the crimson gentle,” she says. By simply coasting, “I’m not burning fuel or electrical energy within the meantime.” If one automotive, resembling an automatic car, slows down on the method to an intersection, then the traditional, non-automated automobiles behind it would even be compelled to decelerate, so the affect of such environment friendly driving can lengthen far past simply the automotive that’s doing it.

    That’s the essential concept behind eco-driving, Wu says. However to determine the affect of such measures, “these are difficult optimization issues” involving many alternative components and parameters, “so there’s a wave of curiosity proper now in tips on how to clear up exhausting management issues utilizing AI.” 

    The brand new benchmark system that Wu and her collaborators developed based mostly on city eco-driving, which they name “IntersectionZoo,” is meant to assist handle a part of that want. The benchmark was described intimately in a paper offered on the 2025 Worldwide Convention on Studying Illustration in Singapore.

    Taking a look at approaches which have been used to handle such advanced issues, Wu says an vital class of strategies is multi-agent deep reinforcement studying (DRL), however an absence of sufficient commonplace benchmarks to guage the outcomes of such strategies has hampered progress within the area.

    The brand new benchmark is meant to handle an vital situation that Wu and her workforce recognized two years in the past, which is that with most present deep reinforcement studying algorithms, when skilled for one particular scenario (e.g., one explicit intersection), the outcome doesn’t stay related when even small modifications are made, resembling including a motorbike lane or altering the timing of a visitors gentle, even when they’re allowed to coach for the modified state of affairs.

    Actually, Wu factors out, this downside of non-generalizability “isn’t distinctive to visitors,” she says. “It goes again down all the way in which to canonical duties that the neighborhood makes use of to guage progress in algorithm design.” However as a result of most such canonical duties don’t contain making modifications, “it’s exhausting to know in case your algorithm is making progress on this type of robustness situation, if we don’t consider for that.”

    Whereas there are lots of benchmarks which can be presently used to guage algorithmic progress in DRL, she says, “this eco-driving downside encompasses a wealthy set of traits which can be vital in fixing real-world issues, particularly from the generalizability standpoint, and that no different benchmark satisfies.” For this reason the 1 million data-driven visitors eventualities in IntersectionZoo uniquely place it to advance the progress in DRL generalizability.  In consequence, “this benchmark provides to the richness of the way to guage deep RL algorithms and progress.”

    And as for the preliminary query about metropolis visitors, one focus of ongoing work shall be making use of this newly developed benchmarking software to handle the actual case of how a lot affect on emissions would come from implementing eco-driving in automated autos in a metropolis, relying on what share of such autos are literally deployed.

    However Wu provides that “slightly than making one thing that may deploy eco-driving at a metropolis scale, the principle objective of this research is to assist the event of general-purpose deep reinforcement studying algorithms, that may be utilized to this utility, but additionally to all these different purposes — autonomous driving, video video games, safety issues, robotics issues, warehousing, classical management issues.”

    Wu provides that “the undertaking’s objective is to supply this as a software for researchers, that’s overtly out there.” IntersectionZoo, and the documentation on tips on how to use it, are freely out there at GitHub.

    Wu is joined on the paper by lead authors Vindula Jayawardana, a graduate scholar in MIT’s Division of Electrical Engineering and Laptop Science (EECS); Baptiste Freydt, a graduate scholar from ETH Zurich; and co-authors Ao Qu, a graduate scholar in transportation; Cameron Hickert, an IDSS graduate scholar; and Zhongxia Yan PhD ’24. 



    Source link

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleQ&A: A roadmap for revolutionizing health care through data-driven innovation | MIT News
    Next Article The CNN That Challenges ViT
    ProfitlyAI
    • Website

    Related Posts

    Artificial Intelligence

    Why Should We Bother with Quantum Computing in ML?

    October 22, 2025
    Artificial Intelligence

    Federated Learning and Custom Aggregation Schemes

    October 22, 2025
    Artificial Intelligence

    Implementing DRIFT Search with Neo4j and LlamaIndex

    October 22, 2025
    Add A Comment
    Leave A Reply Cancel Reply

    Top Posts

    MIT-forskare lär AI att känna igen personliga objekt

    October 17, 2025

    OpenAI’s Unstoppable Growth Streak Just Hit Another Insane Milestone

    August 5, 2025

    Not Everything Needs Automation: 5 Practical AI Agents That Deliver Enterprise Value

    June 6, 2025

    Forget Siri: Elon Musk’s Grok Just Took Over Your iPhone

    April 24, 2025

    How I Fine-Tuned Granite-Vision 2B to Beat a 90B Model — Insights and Lessons Learned

    July 25, 2025
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    Most Popular

    A $100M AI Super PAC Is About to Reshape US Elections

    September 3, 2025

    AI’s impact on the job market: Conflicting signals in the early days

    April 29, 2025

    What you may have missed about Trump’s AI Action Plan

    July 29, 2025
    Our Picks

    Why Should We Bother with Quantum Computing in ML?

    October 22, 2025

    Federated Learning and Custom Aggregation Schemes

    October 22, 2025

    How To Choose The Perfect AI Tool In 2025 » Ofemwire

    October 22, 2025
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About us
    • Contact us
    Copyright © 2025 ProfitlyAI All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.