Using Deep Deterministic Policy Gradient (DDPG) with monocular images, the vehicle learns autonomous lane following within 10 episodes, achieving over 200 meters of safe driving.

Alex Kendall, Jeffrey Hawke, David Janz et al.

TextWorld is a sandbox learning environment for training RL agents on text-based games, supporting handcrafted or automatically generated games.

Marc-Alexandre Côté, Ákos Kádár, Xingdi Yuan et al.

Proposes a self-supervised monocular endoscopy depth estimation method using SfM-generated sparse supervision, achieving submillimeter accuracy.

Xingtong Liu, Ayushi Sinha, Mathias Unberath et al.

Proposed TOG-Net, a framework for task-oriented grasping via simulated self-supervision; achieved 71.1% sweeping and 80.0% hammering success rates.

Kuan Fang, Yuke Zhu, Animesh Garg et al.

DARTS employs differentiable architecture search with bilevel optimization, achieving high efficiency and state-of-the-art results on CIFAR-10 and PTB datasets.

Hanxiao Liu, Karen Simonyan, Yiming Yang

Edgent framework combines DNN partitioning and early-exit to enable device-edge collaborative inference with low latency, validated on Raspberry Pi with significant performance gains.

En Li, Zhi Zhou, Xu Chen

DecaNLP frames ten NLP tasks as question answering; MQAN uses multi-pointer decoder and coattention, achieving state-of-the-art results in transfer, domain adaptation, and zero-shot learning.

Bryan McCann, Nitish Shirish Keskar, Caiming Xiong et al.

Proposes EQL÷, a neural network with division units, for symbolic equation discovery and extrapolation, achieving high accuracy with minimal data.

Subham S. Sahoo, Christoph H. Lampert, Georg Martius

Proposes VirtualHome, a simulation system that models household activities via programs, using deep learning to convert natural language and videos into executable task sequences, advancing robot task understanding.

Xavier Puig, Kevin Ra, Marko Boben et al.

Introduced VoxCeleb2 dataset and ResNet models, achieving state-of-the-art speaker recognition in noisy environments.

Joon Son Chung, Arsha Nagrani, Andrew Zisserman

Proposed SSSP validation protocol combining ∆-factor and performance metrics to develop high-precision and efficient solid-state pseudopotential libraries.

Gianluca Prandini, Antimo Marrazzo, Ivano E. Castelli et al.

Proposed a generative image inpainting system using gated convolution, achieving superior performance on Places2 and CelebA-HQ datasets.

Jiahui Yu, Zhe Lin, Jimei Yang et al.

FPN with ResNet50 for multi-class land segmentation achieved IoU 0.493 on DEEPGLOBE 2018.

Selim S. Seferbekov, Vladimir I. Iglovikov, Alexander V. Buslaev et al.

PipeDream combines pipeline, model, and data parallelism, reducing communication by up to 95% and boosting GPU utilization for large DNN training.

Aaron Harlap, Deepak Narayanan, Amar Phanishayee et al.

Proposed PinSage, combining random walks and graph convolutions, scales to 3 billion nodes and 18 billion edges for web-scale recommendation.

Rex Ying, Ruining He, Kaifeng Chen et al.

Introduces singular value transformation (SVT) algorithm leveraging quantum signal processing, exponentially accelerating matrix functions including Hamiltonian simulation, pseudoinverse, and machine learning.

András Gilyén, Yuan Su, Guang Hao Low et al.

Sample-based Schrödinger bridge using maximum likelihood and importance sampling, suitable for high-dimensional probability transport.

Michele Pavon, Esteban G Tabak, Giulio Trigila

Proposes stepwise relative reachability combined with a new baseline to prevent bad incentives in impact penalties, improving safety in reinforcement learning.

Victoria Krakovna, Laurent Orseau, Ramana Kumar et al.

RedNet employs residual encoder-decoder with depth fusion and pyramid supervision, achieving 47.8% mIoU on SUN RGB-D.

Jindong Jiang, Lunan Zheng, Fei Luo et al.

GamePad integrates deep learning with Coq, predicting proof steps to enhance automation and human-AI collaboration.

Daniel Huang, Prafulla Dhariwal, Dawn Song et al.