AGATHA uses graph-mining and Transformer models for hypothesis generation in medical research, achieving a recommendation accuracy of 0.901.

Justin Sybrandt, Ilya Tyagin, Michael Shtutman et al.

Proposed streaming sparse GP ensemble for implicit surface mapping, resource-efficient and comparable to exact GP.

Johannes A. Stork, Todor Stoyanov

VI-ULCB gives self-play in zero-sum Markov games \tilde{O}(\sqrt{H^3S^2ABT}) regret against fully adaptive opponents.

Yu Bai, Chi Jin

Using importance weighting to correct feedback shift under delayed feedback, enhancing conversion rate prediction accuracy.

Shota Yasui, Gota Morishita, Komei Fujita et al.

Ensemble grammar induction for anomaly detection improves robustness by combining multiple parameter results, achieving near state-of-the-art performance with linear complexity.

Yifeng Gao, Jessica Lin, Constantin Brif

Combines VAE and deep SVDD with ICAD for real-time out-of-distribution detection in CPS.

Feiyang Cai, Xenofon Koutsoukos

This study formulates power-law scaling laws for neural language models, linking model size, data, and compute to performance, guiding efficient training strategies.

Jared Kaplan, Sam McCandlish, Tom Henighan et al.

FixMatch combines consistency regularization and pseudo-labeling, achieving 94.93% accuracy on CIFAR-10 with 250 labels.

Kihyuk Sohn, David Berthelot, Chun-Liang Li et al.

Hierarchical predictor-corrector framework with differentiable PDE solver enables long-term control of nonlinear PDE systems like Navier-Stokes.

Philipp Holl, Vladlen Koltun, Nils Thuerey

Reformer combines reversible residual layers and locality-sensitive hashing (LSH) attention, reducing complexity from O(L^2) to O(L log L) for long sequences.

Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya

This paper establishes near-optimal approximation bounds for deep ReLU networks approximating smooth functions, with errors of O(N^{-2s/d}L^{-2s/d}) when width and depth are optimized.

Jianfeng Lu, Zuowei Shen, Haizhao Yang et al.

SPACE combines spatial attention and scene mixture, enabling scalable unsupervised multi-object scene representation with explicit object and background factors.

Zhixuan Lin, Yi-Fu Wu, Skand Vishwanath Peri et al.

Proposes hybrid reinforcement learning (Hybrid RL) based on MPO to efficiently handle discrete and continuous actions simultaneously, improving robotic control.

Michael Neunert, Abbas Abdolmaleki, Markus Wulfmeier et al.

PAC Confidence Sets combine temperature scaling with a binomial generalization bound, yielding finite-sample coverage on ImageNet, tracking, and Half-Cheetah.

Sangdon Park, Osbert Bastani, Nikolai Matni et al.

OpenAI Five used large-scale deep RL, training 10 months to beat Dota 2 champions, overcoming environment complexity.

OpenAI, :, Christopher Berner et al.

Proposes federated learning algorithms with privacy and efficiency improvements; experiments show 20% accuracy gain on CIFAR-10.

Peter Kairouz, H. Brendan McMahan, Brendan Avent et al.

Deep symbolic regression employs RNNs and risk-seeking policy gradients to recover mathematical expressions, outperforming traditional methods on benchmark problems.

Brenden K. Petersen, Mikel Landajuela, T. Nathan Mundhenk et al.

Introduces ‘Effective Model Complexity’ to explain double descent phenomena, revealing non-monotonic effects of model size and training epochs on generalization.

Preetum Nakkiran, Gal Kaplun, Yamini Bansal et al.

Dreamer leverages latent space imagination with deep models and analytic gradients to achieve efficient long-horizon control on 20 visual tasks, surpassing prior methods.

Danijar Hafner, Timothy Lillicrap, Jimmy Ba et al.

Proposes Continuous Graph Neural Networks (CGNN) using neural ODEs, addressing over-smoothing and capturing long-range dependencies, achieving state-of-the-art node classification accuracy.

Louis-Pascal A. C. Xhonneux, Meng Qu, Jian Tang