iiinova — where your digital self keeps evolving

Reposted by André Biedenkapp

Cathy Wu @cathywu.bsky.social · 05/05/2025

Excited to announce IntersectionZoo, a benchmark that uses a real-world traffic problem to test generalization progress in deep reinforcement learning, particularly multi-agent contextual RL. 🤖🧠 MIT News coverage: news.mit.edu/2025/new-too... Benchmark: intersectionzoo-docs.readthedocs.io

In building IntersectionZoo, three main logical layers are used to abstract functionality as shown in this figure. In the traffic scenario modeling layer, we first build data-driven simulation environments of signalized intersections and then use them to build traffic scenarios at those intersections. Concretely, an intersection is first defined by factors such as lane lengths, lane counts, road grades, turn lane configuration, and speed limit of each approach. Then, vehicle type, age, and fuel type distributions are used with appropriate traffic flow rates and human driven vehicle behaviors to define a realistic traffic flow. Each intersection is then used to define traffic scenarios by further assigning representative atmospheric temperature and humidity values based on the season. Further scenario variations can be achieved by changing the eco-driving adoption level (0%-100%).

Once traffic scenarios are modeled, the CMDP modeling layer is used to contain them as Conextual Markov Decision Process (CMDP) and define the state, action, reward to be used with the reinforcement learning algorithms. CMDP is used to model problem variations. In IntersectionZoo, each city is modeled as a CMDP with each traffic scenario stemming from the city as a problem variation. In a CMDP, each problem variation is an MDP defined by a problem variation context and called a context-MDP. In eco-driving, those context-MDPs are defined by the following state, action and reward functions and formulate as a multi-agent control problem.

In user configuration layer, we provide users the flexibility to configure their experimental setup.

An example of cooperative eco-driving at signalized intersections where CVs are controlled to minimize the total exhaust emissions of the fleet (both CVs and HDVs) while minimizing the impact on the travel time of each vehicle. While CVs are controllable by some defined eco-driving strategy, HDVs are driven by humans and can not be controlled. However, CVs implicitly control HDVs through car-following dynamics and by forming locally cooperative teams for better system control (by controlling the oppotunities of HDVs to overake CVs).

we show the intersection distribution of each city and the distributions of intersection features of each city. The full dataset of 16,334 intersections in compliance with SUMO simulator can be found in the [here](https://drive.google.com/drive/folders/1y3W83MPfnt9mSFGbg8L9TLHTXElXvcHs?usp=sharing).

1183

Reposted by André Biedenkapp

AutoML Conference @automl-conf.bsky.social · 22/04/2025

The 5th NAS Unseen-Data Challenge at #AutoML25 is on! 🥊 How well do your #NAS pipelines perform on completely new, unseen datasets? 📊 Phase 1 is open until June - grab the starter kit and start building! Winners will be announced at #AutoMLConf in #NYC 🗽 More info 👉 www.nascompetition.com

nascompetition.com

NAS Unseen-Data

054

Reposted by André Biedenkapp

AutoML Conference @automl-conf.bsky.social · 13/03/2025

The #AutoMLConf deadline was extended! We’re looking for groundbreaking contributions in #AutoML, NAS, LLMs, agent-based systems, hyperparameter optimization, benchmarking, algorithm selection, and more. Don’t miss your chance to showcase your work in #NYC! 📅 Deadline: March 31 🌐 CfP: 2025.automl.cc

2025.automl.cc

AutoML

085

Reposted by André Biedenkapp

Katharina Eggensperger @keggensperger.bsky.social · 11/03/2025

📣 The 5th AutoML School is coming to Tübingen! Whether you are new to #AutoML or an experienced researcher, this is the Summer School for you. Join us for 4 exciting days exploring basic topics, studying cutting edge-research and having fun. 📅 June 10th-13th, Tübingen, Germany 🔗 automlschool.org

automlschool.org

4137

Reposted by André Biedenkapp

Carola Doerr @caroladoerr.bsky.social · 04/03/2025

Job alert: full-time researcher position for AI with CNRS here in France. Initial contract for 5 years (tenure track). Attractive starting package and better salary than regular positions. I have been w/ @cnrs.fr since 2013, it's great! Applications due 31/03 www.ins2i.cnrs.fr/en/cnrsinfo/...

ins2i.cnrs.fr

Choose France - CNRS AI Rising Talents

As part of the French Strategy in Artificial Intelligence (AI), the Choose France - CNRS AI Rising Talents program offers exceptionally tal

12121

Reposted by André Biedenkapp

Theresa Eimer @theeimer.bsky.social · 16/01/2025

Starting the year off with some AutoRL inspiration: the autorl.org team put together our favorite papers on how to make RL work that we found in 2024 - and it's been a pretty great year! 🌟🚀 Let us know what we missed! autorl.org/blog/retrosp...

autorl.org

2024 in AutoRL

TL;DR: From integrating RL with VLMs and LLMs to hyperparameter tuning, environment design, and generalization, 2024 was packed with innovation. We’ve highlighted top advancements in AutoRL and inclu...

1102

André Biedenkapp @abiedenkapp.bsky.social · 05/12/2024

I've been instructed by @saiprasanna.in to post more. But I'm lazy and it off ideas, so I'll just post my Spotify wrapped 😅 Apparently I made it into the top 2% of Linkin park listeners this year. open.spotify.com/wrapped/shar...

open.spotify.com

Mein Wrapped 2024 – hol dir deins

2024 Wrapped

220

Reposted by André Biedenkapp

AutoML Conference @automl-conf.bsky.social · 05/12/2024

Big news! The AutoML Conference is back! 🎉 Next year, we’re heading to the city that never sleeps: New York City🗽. Save the date: Sept 8–11, 2025. Stay tuned for updates, and in the meantime, check out our website: 2025.automl.cc. See you there? #AutoML25 #AutoMLConf #NYC #AutoML

2025.automl.cc

AutoML

0195

Reposted by André Biedenkapp

Nezihe Merve Gürel @nmervegurel.bsky.social · 03/12/2024

The list of accepted workshops for ICLR 2025 is available at openreview.net/group?id=ICL... @iclr-conf.bsky.social We received 120 wonderful proposals, with 40 selected as workshops.

openreview.net

ICLR 2025 Workshop Proposals

Welcome to the OpenReview homepage for ICLR 2025 Workshop Proposals

15715

Reposted by André Biedenkapp

Sai Prasanna @saiprasanna.in · 23/11/2024

Our paper "Dreaming of Many Worlds" studied zero-shot generalization in world models for Contextual RL and was later accepted by RLC and EWRL. Here's a thread summarizing the work and putting it in the context in retrospect as ideas for potential future collaborations! arxiv.org/abs/2403.10967

arxiv.org

Dreaming of Many Worlds: Learning Contextual World Models Aids Zero-Shot Generalization

Zero-shot generalization (ZSG) to unseen dynamics is a major challenge for creating generally capable embodied agents. To address the broader challenge, we start with the simpler setting of contextual...

3638