Journal of Machine Learning Research: "On the Complexity of Approximating ...| Robocrunch AI

"On the Complexity of Approximating Multimarginal Optimal Transport", by Tianyi Lin, Nhat Ho, Marco Cuturi, Michael I. Jordan. https://t.co/VtMdbzbIfI
— Journal of Machine Learning Research (@JmlrOrg) May 13, 2022

Sequence-to-sequence models are incapable of compositional generalization to inputs that require novel mixing and matching of syntactic structures (Weißenhorn, Yao, Donatelli, @alkoller): https://t.co/j2PZJnjBT9 (h/t @najoungkim)

Shared by Tal Linzen at 2/28/2022

"Scaling-Translation-Equivariant Networks with Decomposed Convolutional Filters", by Wei Zhu, Qiang Qiu, Robert Calderbank, Guillermo Sapiro, Xiuyuan Cheng. https://t.co/t5t3A27x7i

Shared by Journal of Machine Learning Research at 5/13/2022

Short #SIGIR2022 paper: "Faster Learned Sparse Retrieval with Guided Traversal" by @antonio_mallia, Joel Mackenzie, Torsten Suel, and @ntonellotto explores an efficient single-pass traversal algorithm for accelerating DeepImpact. @SIGIRConf

Shared by ielab at 4/1/2022

We used T5X and SeqIO to train PaLM 🌴, and also released a short paper on this infrastructure alongside the model. Check it out at https://t.co/V795jjwl22. Huge thanks to the many contributors and my co-lead authors @hwchung27 @anselmlevskaya @mishra_gaurav_ @jekbradbury!

Shared by Adam Roberts at 4/4/2022

"Stacking for Non-mixing Bayesian Computations: The Curse and Blessing of Multimodal Posteriors" with @YulingYao and @StatModeling has now appeared in JMLR 23(79):1−45, 2022 https://t.co/u9RTYLtdZS

Shared by Aki Vehtari at 5/9/2022

For me this is part of an important long-term effort on computational unification and efficiency in #SpatialAI, prototyping highly parallel algorithms which could be brought "close to the pixel". See https://t.co/0MjvQfnoru for more discussion, and https://t.co/DYgeZVaBHO on GBP.

Shared by Andrew Davison at 4/19/2022

768