If you're interested in GNNs for combinatorial tasks (certainly an exciting time!), we've released our 43-page comprehensive survey on the area! + detailed blueprint of algorithmic reasoning in S3.3. arxiv.org/abs/2102.09544 with Christopher Morris @ ICML Andrea Lodi Elias Khalil qcappart & Didier

A short version of our GNN+CO survey was just accepted by IJCAIconf #IJCAI2021 in the survey track! Looking forward to presenting our work in (virtual) Montréal ✨

Thanks to qcappart and our great team from both Polytechnique (MTL and Paris), we are finally releasing SeaPearl, our open-source RL-guided CP solver (CP-RL) in Julia. All details are available on our new website. corail-research.github.io

[1/3] The Machine Learning for Combinatorial Optimization (ML4CO) NeurIPS 2021 competition aims at improving state-of-the-art combinatorial optimization solvers by replacing / integrating key heuristic components with machine learning models.

Brand-new #NeurIPS2021 #competition about "Machine Learning for Discrete Optimization" - see below for more. Please RT so that we can reach a broad set of people. pokutta.com/blog/news/2021… #ml #ai MONTREAL.AI #optimization #integerprogramming #ml4co

New MSc, Ph.D., and postdoc positions are opening at the SCALE-AI Chair on Data-Driven Supply Chains at Polytechnique Mtl. Research topics on #ORMS, #optimization and interpretable #MachineLearning. Can you relay this announcement to your network? w1.cirrelt.ca/~vidalt/resour…

🎄 A new winter holiday longread - what happened in the Graph ML field in 2021? What to expect in 2022? We reflect upon a dozen of research trends - brought by yours truly, @epsiloncorrect and @SergeyI49013776 (and a surprise cameo appearance in the end) mgalkin.medium.com/graph-ml-in-20…

If you are interested in ML for combinatorial optimization, checkout out our short write-up of the results of the ML4CO NeurIPS competition: arxiv.org/abs/2203.02433. It also includes summaries of the winning approaches.

Call for applications: new MSc, PhD, and postdoc positions opening at the SCALE-AI chair in Polytechnique Mtl! We conduct research on hybrid OR/ML algorithms and interpretable #MachineLearning. Can you relay this announcement to your network? w1.cirrelt.ca/~vidalt/resour…

Are you looking for a summer school on the integration of constraint programming and machine learning? See all details of our ACP summer school: bit.ly/41dWVzT 📅 When? 10-14/07/23 🏰 Where? Leuven, Belgium ✍️Registration: bit.ly/3UJfRUP

I am simultaneously excited & exhausted to announce that, 2 years later, our survey on GNNs for CO has been published Journal of Machine Learning Research (now at 61 pages of length!!!) 😅🚀 jmlr.org/papers/v24/21-… Persistence paid off, guys! 😊 Christopher Morris Andrea Lodi Elias Khalil qcappart Didier Chételat

How capable are web agents at solving knowledge work tasks? 🤔 Are LLMs up to the challenge? 🤖 Introducing WorkArena: a benchmark where agents meet the world 𝘸𝘪𝘭𝘥 web of enterprise software 🌐🖥️ Paper: bit.ly/4a7FiFV Website: bit.ly/3VkdJ87 🧵 1/7

We’re really excited to release this large collaborative work for unifying web agent benchmarks under the same roof. In this TMLR paper, we dive in-depth into #BrowserGym and #AgentLab. We also present some unexpected performances from Claude 3.5-Sonnet