Andrew Rowan
@arowan_mlBayesian reasoning and probabilistic programming in industry.
Similar User
@jesfrellsen
@dustinvtran
@syhw
@ropeharz
@markvanderwilk
@caglarml
@edaxberger
@jameshensman
@jhhhuggins
@dianarycai
@davidobarber
@danielewworrall
@betanalpha
@JCornebise
@_hylandSL
📢 Fascinated to announce our #ICML Workshop on “Counterfactuals in Minds and Machines”! Submit your latest empirical, modeling, computational, and position work on counterfactual reasoning by June 2. CFP on the website: tinyurl.com/27r996fu Speakers and co-organizers in 🧵
New paper: On the unreasonable effectiveness of LLMs for causal inference. GPT4 achieves new SoTA on a wide range of causal tasks: graph discovery (97%, 13 pts gain), counterfactual reasoning (92%, 20 pts gain) & actual causality. How is this possible?🧵 arxiv.org/abs/2305.00050
Very excited for our October 20-22 workshop "Philosophy of Science Meets Machine Learning", hosted by @ml4science Speakers include @cameronjbuckner @FeestUljana @ds_wats0n @uhlily @zacharylipton @KathleenACreel and many more. For the full program, see sites.google.com/view/philml202…
Our 2012 paper ‘On causal and anticausal learning’ just received a Test of Time Honorable Mention at @icmlconf #ICML2022: icml.cc/2012/papers/62…. I am really grateful, and would like to use this occasion for some thoughts on causality and machine learning:
Getting AI to avoid harmful actions by asking why. In collaboration with DT (@babylonhealth) & Rory (@CoMind_ ).
New work from our safety team proposes counterfactual reasoning as a key ingredient for safe and ethical AI. Read more: dpmd.ai/counterfactual… Work by @jonathanrichens, Rory Beard (@CoMind_) and Daniel Thompson (Babylon).
New paper from the @awscloud Causal Representation Learning team and @EPFL_en: "Score matching enables causal discovery of nonlinear additive noise models": the gradient of the log likelihood gives the topological order. arxiv.org/abs/2203.04413
The WHY-21 workshop "Causal Inference & Machine Learning: Why now?" is currently accepting submissions, why21.causalai.net. Our goal is to bring CI & ML researchers together to discuss the nextgen AI! (joint w/ @yudapearl, Y. Bengio, T. Sejnowski, @bschoelkopf) @NeurIPSConf
🚨 9 days left to submit to the @icmlconf workshop on the *Neglected Assumptions in Causal Inference* (sites.google.com/view/naci2021/…). We are eager to hear from researchers inside and outside of computer science! @LauraBBalzer @alexdamour @uhlily @raziehnabi @ShalitUri @ICML2021
In ML, spurious correlations are know-it-when-you-see-it; e.g., changing “Alice rules!” to “Bob rules!” changes predicted sentiment. Such ad-hoc tests are intuitive, but hard to connect to standard practice. Turns out, causality has a lot to say! arxiv.org/abs/2106.00545
Towards Causal Representation Learning: led by @bschoelkopf and myself, with amazing co-authors Stefan Bauer, @rosemary_ke, @NalKalchbrenner, @anirudhg9119, Yoshua Bengio, accepted in the Proceedings of the IEEE. Link: arxiv.org/abs/2102.11107
Underspecification Presents Challenges for Credibility in Modern Machine Learning. (arXiv:2011.03395v1 [cs.LG]) ift.tt/2UdTHv7
Of course Data-driven ML wont be replaced. As Rung-1 in the Ladder, it is an important component in any Causal Inference Engine. It will transform, however, from a data-exclusive engine to a hybrid {data,model}-driven engine. As to far-fetched futures, see causality.cs.ucla.edu/blog/index.php…
It’s unlikely that data-driven ML will be replaced. Rather it will be augmented by causal modeling. And I bet that even causal modeling will eventually become data-driven itself in the form of causal generative models.
We (@anirudhg9119 @jovana_mitr @theophaneweber @janexwang @DaniloJRezende @csilviavr, Stefan Bauer) will be organizing the Causal learning for decision making workshop at #ICLRL2020, sites.google.com/view/causal-le…
#ICLRL2020 will be home to 15 workshops, all run virtually! The workshop organizers are working incredibly hard to bring everything together for 26 April 🤩. See all the amazing workshops we have here👉🏾 medium.com/@iclr_conf/vir…
"absolutely brilliant" —Nobel Laureate Danny Kahneman The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence It's what I wish I had had time to say at the #AIDebate :) Finally ready, free, on arXiv. Happy Reading! arxiv.org/abs/2002.06177
Delighted to announce the CFA for our next Kinds of Intelligence conference, June 22-25 in Cambridge. The event aims to explore mind and cognition with an emphasis on artificial and non-human biological intelligence. Please share widely! philevents.org/event/show/810…
The case for Bayesian Deep Learning – succinctly and crisply argued by @andrewgwils arxiv.org/abs/2001.10995
Mini thread: If you haven't already @FChollet's beautiful & insightful paper on intelligence & AI, you should. An elegant distillation of where we are now, & an intriguing proposal for how to make progress. arxiv.org/abs/1911.01547
Check out our extensive review paper on normalizing flows! This paper is the product of years of thinking about flows: it contains everything we know about them, and many new insights. With @eric_nalisnick, @DeepSpiker, @shakir_za, @balajiln arxiv.org/abs/1912.02762 Thread 👇
Looking for something to read in your flight to #NeurIPS2019? Read about Normalizing Flows from our extensive review paper (also with new insights on how to think about and derive new flows) arxiv.org/abs/1912.02762 with @gpapamak @eric_nalisnick @DeepSpiker @balajiln @shakir_za
United States Trends
- 1. Colby 26,9 B posts
- 2. #UFCTampa 55,7 B posts
- 3. Buckley 17,9 B posts
- 4. Cub Swanson 6.092 posts
- 5. Jeanty 72 B posts
- 6. Heisman 158 B posts
- 7. Travis Hunter 126 B posts
- 8. #SNME 176 B posts
- 9. Munguia 7.433 posts
- 10. Canelo 2.688 posts
- 11. Kape 7.894 posts
- 12. Hulk Hogan 3.732 posts
- 13. #UFCFightNight 1.373 posts
- 14. Usman 16,6 B posts
- 15. #INDvsAUS 41 B posts
- 16. Billy Q N/A
- 17. Deion 14,5 B posts
- 18. #GCWHigh 1.227 posts
- 19. Petrino 2.976 posts
- 20. Jacoby 4.487 posts
Who to follow
-
Jes Frellsen
@jesfrellsen -
Dustin Tran
@dustinvtran -
Gabriel Synnaeve
@syhw -
Robert Peharz
@ropeharz -
Mark van der Wilk
@markvanderwilk -
Caglar Gulcehre @ NeuRIPS Vancouver
@caglarml -
Erik Daxberger
@edaxberger -
James Hensman
@jameshensman -
Jonathan Huggins
@jhhhuggins -
Diana Cai
@dianarycai -
David Barber
@davidobarber -
Daniel Worrall
@danielewworrall -
\mathfrak{Michael "Shapes Dude" Betancourt}
@betanalpha -
Julien Cornebise
@JCornebise -
_hylandSL
@_hylandSL
Something went wrong.
Something went wrong.