CausalPFN: Amortized Causal Effect Estimation via In-Context Learning: arxiv.org/abs/2506.07918

Can neural networks learn to map from observational datasets directly onto causal effects? YES! Introducing CausalPFN, a foundation model trained on simulated data that learns to do in-context heterogeneous causal effect estimation, based on prior-fitted networks (PFNs). Joint work with @Layer6AI & @hamid_R_kamkar w/ @_valthomas, Jeremy Ma, Benson Li, Jesse C. Cresswell, & @rahulgk 📝ArXiv: arxiv.org/abs/2506.07918 🔗Code: github.com/vdblm/CausalPFN/ 🗣️Oral paper @ ICML SIM workshop 🧵[1/7]

🚀 #ICLR2025 & #CHI2025 are just around the corner — and we’re excited to welcome you to our Bidirectional 👫Human-AI🤖 Alignment events! 🏅 "Golden Sponsors"🏅 A heartfelt thank you to our two generous Golden Sponsors: 🌟@Prolific and 🌟@Layer6AI of @TDBank_US! Their support is making it possible for us to fund 13 travel grant recipients and 4-5 paper awards — stay tuned for our awards reveal at our #ICLR event! 🎤 Speaker Lineup🎤 We’re thrilled to welcome our amazing speakers: 🧠 In-person at ICLR: @_beenkim , @HungyiLee2, @Pavel_Izmailov, @RichardMCNgo 🧠 In-person at CHI: @bradamyers 🧠 Virtual via Zoom: @fraukolos, Dan Bohus 👏 Warm welcome to Tammy Masterson (Technical Partnerships Lead, @AISecurityInst, UK Government), joining us in person for our panel! 🧑‍💻 Organizer Team & Schedule🧑‍💻 Catch us in person at: ✨ ICLR: @ABosselut, @SLED_AI, @yangli169, and me ✨ CHI: @tknearem, @tongshuangwu, @lxieyang, @YunHuang_HCI ✨ Virtual host: @tanmit, Marti Hearst, @ziqiao_ma, @reshmigh 📅 Schedule & details: bialign-workshop.github.io/#…. Our all-day workshop will be streamed via Zoom, and virtual poster sessions via Slack. Links soon on our website! 📜 Accepted Papers📜 From over 100 submissions, we accepted 72 incredible papers! 📢 10 Orals, 62 Posters, 5 Tiny Papers 🏅Award winners will be announced during the workshop! 📚 Browse the full paper list here: bialign-workshop.github.io/#… 🎉 Can’t wait to connect and explore bidirectional human-AI alignment with you at #ICLR and #CHI! @iclr_conf @acm_chi @uw_ischool @uwcse @michigan_AI #BiAlign #HumanAIAlignment #ICLR #CHI✨

Our paper on Text-To-SQL translation accepted at #NAACL2024 has been selected for an oral presentation!🎉 🚀We introduce MSc-SQL, an open-source method that  achieves state-of-the-art performance among small language models. arxiv.org/abs/2410.12916 github.com/layer6ai-labs/msc…

New Blog Alert! Parallel processing in Python can supercharge performance, but it also comes with challenges—especially memory errors that can be tricky to diagnose and fix. In this blog post, we discuss how to avoid common pitfalls. layer6.ai/troubleshooting-me…

Our team will be attending #NeurIPS2024 in Vancouver! Our scientists will be presenting research at poster sessions and workshops—don’t miss the chance to connect with us and explore our latest findings. Let’s shape the future of ML together!

Choosing the right data matters! 🌟 Learn how CCDR enhances object detection by aligning source & target data with smart clustering & pruning. Boost efficiency & performance with better datasets. 🔍📊 Read more: layer6.ai/smart-dataset-sele… #AI #MachineLearning #DataScience

🚀 We’re hiring! Join our team at the intersection of machine learning & finance. We're looking for: 🔹 Software Engineers 🔹 Product Owners 🔹 Research Scientists Help us innovate in the financial industry with cutting-edge tech! See our open positions at layer6.ai/careers/

Feeling fresh! Our website just got a makeover, and we’re thrilled to unveil our brand-new blog section! Dive into exciting content about our research, engineering, and product insights in machine learning. Check it out and let us know what you think! layer6.ai/

🚀 Exciting News! 🚀 We are thrilled to announce that two of our research papers have been accepted to NeurIPS 2024 including one 💡 Spotlight! 💡 Stay tuned for more updates as we head to NeurIPS in December! 🙌 Papers: arxiv.org/abs/2406.03537 arxiv.org/abs/2406.05207

Layer 6 was onstage for this year's premier banking conference held in Seattle! Jesse Cresswell spoke about the future of #GenAI in banking and how it will transform the way we work, but also how to manage emerging risks by integrating #Trustworthy AI practices into applications.

Super proud of the @Layer6AI team, winning @recsyschallenge, besting top global tech companies and grad schools. (2016 winner was Alibaba) x.com/recsyschallenge/status…

We won!... x.com/recsyschallenge/status…

Seizing the #AI future. @globeandmail op ed I wrote w Geoff Hinton, @tjpoutanen, Rich Zemel & Ed Clark. theglobeandmail.com/report-o…

Inclusion here makes our moms proud! Thx @shivon @jamescham @BloombergBeta @HarvardBiz. x.com/roybahat/status/793915…

And we're back. Kicking off this session is Tomi Poutanen, who will give us some deep learning advice for TV recommendations. #RecSysTV

Using #deeplearning @Layer6AI obtain more accurate models, less complex feature spaces, and is able to have many more inputs for #recsys2016

We have @comcast @verizon @netflix and @Layer6AI in our panel - the future and present of #TV #recsys2016