Hacker News
- Turing Complete Transformers: Two Transformers Are More Powerful Than One https://openreview.net/forum?id=MGWsPGogLH 77 comments
- Turing complete transformers: Two transformers are more powerful than one https://openreview.net/pdf?id=MGWsPGogLH 3 comments
- Predictive Coding Approximates Backprop Along Arbitrary Computation Graphs https://openreview.net/forum?id=PdauS7wZBfC 12 comments
- Learning to Design RNA https://openreview.net/forum?id=ByfyHh05tQ 2 comments
- Winner’s Curse? On pace, progress, and empirical rigor [pdf] https://openreview.net/pdf?id=rJWF0Fywf 4 comments
- A computational model of the Moth Olfactory Network learns to read MNIST [pdf] https://openreview.net/pdf?id=HyYuqoCUz 14 comments
- [D] MetaGPT grossly misreported baseline numbers and got an ICLR Oral! https://openreview.net/forum?id=VtmBAGCN7o 38 comments machinelearning
- MuDreamer: Learning Predictive World Models without Reconstruction https://openreview.net/forum?id=9pe38WpsbX 2 comments reinforcementlearning
- [D] First ICLR submission. Got rejected. Great experience overall. https://openreview.net/forum?id=4fbFKO4a2W 7 comments machinelearning
- [R] Continual Learning : Applications and the Road Forward ( Nov.30 2023 ) https://openreview.net/pdf/11ccf96a5da7d057f02f2287c5cc1f2a2296842b.pdf 2 comments machinelearning
- [R] Distort, Distract, Decode: Instruction-Tuned Model Can Refine its Response from Noisy Instructions https://openreview.net/forum?id=IqJ3CU3flr 2 comments machinelearning
- [R] Mamba: Linear-Time Sequence Modeling with Selective State Spaces https://openreview.net/forum?id=AL1fq05o7H 4 comments machinelearning
- [D] Score 4.5 GNN paper from Muhan Zhang at Peking University was amazingly accepted by NeurIPS 2022 https://openreview.net/forum?id=nN3aVRQsxGd 17 comments machinelearning
- Relative representations enable zero-shot latent space communication https://openreview.net/forum?id=SrC-nwieGJ 29 comments machinelearning
- [R] ZerO Initialization: Initializing Neural Networks with only Zeros and Ones https://openreview.net/forum?id=1AxQpKmiTc 45 comments machinelearning
- What's the intuition behind the reward scaling with discounted return statistics? https://openreview.net/attachment?id=r1etN1rtPB&name=original_pdf 11 comments reinforcementlearning
- [D] Git Re-Basin Paper Accused of Misinformation https://openreview.net/forum?id=CQsmMYmlP5T¬eId=9liIVMeFFnW 22 comments machinelearning
- Self-Programming Artificial Intelligence https://openreview.net/forum?id=SKat5ZX5RET 13 comments technology
- [D] Data Augmentation in Transformer feature space? (Master Thesis) https://openreview.net/pdf?id=HJ9rLLcxg 2 comments machinelearning
- Data Augmentation in Transformer feature space? https://openreview.net/pdf?id=HJ9rLLcxg 10 comments deeplearning
- "Learning Behaviors through Physics-driven Latent Imagination", Richard et al 2021 (Dreamer for boat/drone) https://openreview.net/forum?id=YwDvofEWlEx 4 comments reinforcementlearning
- "Boosting Search Engines with Interactive Agents", Ciaramita et al 2022 {G} (MuZero & Decision-Transformer T5 for sequences of queries) https://openreview.net/forum?id=0ZbPmmB61g#google 3 comments reinforcementlearning
- [R] Understanding Dimensional Collapse in Contrastive Self-supervised Learning https://openreview.net/forum?id=YevsQ05DEN7 2 comments machinelearning
- [R] PyTorch Implementation of the Natural Posterior Network https://openreview.net/forum?id=tV3N0DWMxCg 7 comments machinelearning
- CFR poker solvers will be replaced by Lazy CFR solvers in the near future? https://openreview.net/forum?id=rJx4p3NYDB 13 comments poker
- [D] Questions about possible self-plagiarism in this NeurIPS paper. https://openreview.net/forum?id=z1F9G4VnGZ-#all 4 comments machinelearning
- [R] AI and the Everything in the Whole Wide World Benchmark https://openreview.net/pdf?id=j6NxpQbREA1 18 comments machinelearning
- Fine-tuning BERT using labelled data to get sentence embeddings https://openreview.net/pdf?id=OHAIVOOl7Vl 6 comments learnmachinelearning
- "Transformers are Meta-Reinforcement Learners", Anonymous 2021 https://openreview.net/forum?id=H7Edu1_IZgR 4 comments reinforcementlearning
- [R] Patches Are All You Need? https://openreview.net/forum?id=tvhs5y4dnvm 11 comments machinelearning
- [R] Scaling Laws vs Model Architectures: How does Inductive Bias Influence Scaling? An Extensive Empirical Study on Language Tasks https://openreview.net/forum?id=wrtp36cbl61 3 comments machinelearning
- [R] Can a Fruit Fly Learn Word Embeddings? https://openreview.net/pdf?id=xfmsoxdxfcg 2 comments machinelearning
- Why the “RL as Inference” proposal is wrong, or how to properly connect RL with Bayesian Inference https://openreview.net/forum?id=S1xitgHtvS 6 comments reinforcementlearning
- Deep Learning for Symbolic Mathematics https://openreview.net/pdf?id=s1ezyehfds 7 comments math
- "StrokeNet: A Neural Painting Environment", Zheng et al 2018 [SPIRAL with a learned/differentiable painting 'world model'] https://openreview.net/forum?id=HJxwDiActX 4 comments reinforcementlearning
- [R] Exploration by random distillation (predicting outputs of a random network) (new Sota on Montezuma) https://openreview.net/forum?id=H1lJJnR5Ym 9 comments reinforcementlearning
- "R2D2: Recurrent Experience Replay in Distributed Reinforcement Learning", Anonymous 2018 [new ALE/DMLab-30 SOTA: "exceeds human-level in 52/57 ALE"; large improvement over Ape-X using a RNN] https://openreview.net/forum?id=r1lyTjAqYX 4 comments reinforcementlearning
- Distributed Distributional Deep Deterministic Policy [R] Gradient [D4PG] (DPG + N-step + prioritized replay) get state of the art performance https://openreview.net/forum?amp%3BnoteId=SyZipzbCb&id=SyZipzbCb 5 comments reinforcementlearning
- "Distributed Prioritized Experience Replay [Ape-X DQN/Ape-X DPG]", Anonymous 2017 (434% median human performance; 2.5k on Montezuma's Revenge) https://openreview.net/forum?amp%3BnoteId=H1Dy---0Z&id=H1Dy---0Z 6 comments reinforcementlearning
- Deep Learning research from the University of Oxford and Google DeepMind can accurately deduce sentences from visual analysis of - LipNet achieves 93.4% accuracy http://openreview.net/pdf?id=bkjlksqxg 12 comments technology
Linking pages
- GitHub - dive-into-machine-learning/dive-into-machine-learning: Free ways to dive into machine learning with Python and Jupyter Notebook. Notebooks, courses, and other links. (First posted in 2016.) https://github.com/hangtwenty/dive-into-machine-learning 39 comments
- Keeping a Bibliography https://bastian.rieck.me/blog/posts/2020/bibliography/ 32 comments
- Call for Main Conference Papers - EMNLP 2024 https://2024.emnlp.org/calls/main_conference_papers/#presentation-at-the-conference 8 comments
- Reviewer Instructions | ECCV 2020 https://eccv2020.eu/reviewer-instructions/ 7 comments
- How to keep up with the latest research and trends in ML | by Igor Susmelj | Towards Data Science https://towardsdatascience.com/how-to-keep-up-with-the-latest-research-and-trends-in-ml-a45a356b1001?amp%3Bsk=15dd22b79619b83b7e1313025c9f8a27&source=friends_link 4 comments
- Open Peer Review Considered Harmful? https://bastian.rieck.me/blog/posts/2023/peer_review/ 2 comments
- A short analysis of ICLR 2020 reviews https://bastian.rieck.me/blog/posts/2019/iclr_analysis/ 0 comments
- How to keep up with the latest research and trends in ML | by Igor Susmelj | Towards Data Science https://towardsdatascience.com/how-to-keep-up-with-the-latest-research-and-trends-in-ml-a45a356b1001?sk=15dd22b79619b83b7e1313025c9f8a27&source=friends_link 0 comments
- ICLR 2020 | A Look at Three Interesting Papers on the Robustness of Neural Networks | by Synced | SyncedReview | Medium https://medium.com/syncedreview/iclr-2020-a-look-at-three-interesting-papers-on-the-robustness-of-neural-networks-689f88a5594e 0 comments
- How to keep up with the latest research and trends in ML | by Igor Susmelj | Towards Data Science https://towardsdatascience.com/how-to-keep-up-with-the-latest-research-and-trends-in-ml-a45a356b1001 0 comments
- The Reproducibility Challenge as an Educational Tool | by Jesse Dodge | PapersWithCode | Medium https://medium.com/paperswithcode/the-reproducibility-challenge-as-an-educational-tool-cd1596e3716c 0 comments
- GitHub - dipakkr/A-to-Z-Resources-for-Students: ✅ Curated list of resources for college students https://github.com/dipakkr/A-to-Z-Resources-for-Students 0 comments
- Rewiewing is a Contract https://bastian.rieck.me/blog/posts/2023/contract/ 0 comments