Proposed DrQA combines efficient retrieval and deep reading, achieving 70%+ accuracy on SQuAD and outperforming baselines.

Danqi Chen, Adam Fisch, Jason Weston et al.

ProcNets leverages a large-scale YouCook2 dataset to achieve unsupervised procedure segmentation in long videos, outperforming baselines with 52% Jaccard and 48% mIoU.

Luowei Zhou, Chenliang Xu, Jason J. Corso

Deep RL-based socially aware motion planning enables autonomous robots to navigate safely and naturally among pedestrians, respecting social norms like passing on the right.

Yu Fan Chen, Michael Everett, Miao Liu et al.

Deep appearance features integrated into SORT reduce identity switches by 45%, enhancing long-term occlusion tracking.

Nicolai Wojke, Alex Bewley, Dietrich Paulus

Proposes importance sampling-based coreset construction, optimizing data reduction for k-means and other ML tasks with theoretical guarantees.

Olivier Bachem, Mario Lucic, Andreas Krause

BOCA algorithm supports continuous multi-fidelity spaces, significantly improving optimization efficiency.

Kirthevasan Kandasamy, Gautam Dasarathy, Jeff Schneider et al.

Introduces R-GCN for knowledge graph completion, achieving 29.8% improvement in link prediction.

Michael Schlichtkrull, Thomas N. Kipf, Peter Bloem et al.

Prototypical Networks utilize class means in an embedding space with Euclidean distance for few-shot classification, achieving state-of-the-art results.

Jake Snell, Kevin Swersky, Richard S. Zemel

Proposes FastQA, a simple neural extractive QA model using question word awareness and RNN, achieving 78.9% F1 on SQuAD.

Dirk Weissenborn, Georg Wiese, Laura Seiffe

Proposes Hypergradient Descent for dynamic learning rate adjustment, reducing manual tuning by leveraging automatic differentiation.

Atilim Gunes Baydin, Robert Cornish, David Martinez Rubio et al.

DeepFM combines FM and DNN for end-to-end CTR prediction, capturing both low- and high-order feature interactions.

Huifeng Guo, Ruiming Tang, Yunming Ye et al.

Proposed a deep learning-based image matting algorithm, achieving state-of-the-art alpha matte accuracy in complex scenes.

Ning Xu, Brian Price, Scott Cohen et al.

This paper introduces Evolution Strategies (ES) as a scalable alternative to deep RL, achieving near real-time training on thousands of cores with minimal communication.

Tim Salimans, Jonathan Ho, Xi Chen et al.

Proposes generative model-based compressed sensing; if G is L-Lipschitz, O(k log L) Gaussian measurements suffice for near-perfect recovery.

Ashish Bora, Ajil Jalal, Eric Price et al.

Proposes a structured self-attentive sentence embedding using a 2D matrix, improving multi-task performance with interpretability.

Zhouhan Lin, Minwei Feng, Cicero Nogueira dos Santos et al.

Gaussian Process-based multiresolution mapping combined with informative path planning reduces agricultural monitoring error by 45%.

Marija Popovic, Teresa Vidal-Calleja, Gregory Hitz et al.

Proposed Global Convolutional Network (GCN) improves semantic segmentation, achieving 82.2% on VOC2012 and 76.9% on Cityscapes.

Chao Peng, Xiangyu Zhang, Gang Yu et al.

DEL-based epistemic planning models knowledge dynamics in multi-agent systems, improving success rates by 20% in complex scenarios.

Thomas Bolander

Combining self-supervised learning and imitation, the robot successfully manipulates ropes using 60K interaction data.

Ashvin Nair, Dian Chen, Pulkit Agrawal et al.

Training deep spiking neural networks on BrainScaleS wafer-scale hardware using in-the-loop backpropagation, achieving ~95% accuracy from initial 72%.

Sebastian Schmitt, Johann Klaehn, Guillaume Bellec et al.