NeurIPS 2018

Data-Efficient Hierarchical Reinforcement Learning

NeurIPS 2018 tensorflow/models

In this paper, we study how we can develop HRL algorithms that are general, in that they do not make onerous additional assumptions beyond standard RL algorithms, and efficient, in the sense that they can be used with modest numbers of interaction samples, making them suitable for real-world problems such as robotic control.

HIERARCHICAL REINFORCEMENT LEARNING

Data-Efficient Hierarchical Reinforcement Learning

NeurIPS 2018 tensorflow/models

In this paper, we study how we can develop HRL algorithms that are general, in that they do not make onerous additional assumptions beyond standard RL algorithms, and efficient, in the sense that they can be used with modest numbers of interaction samples, making them suitable for real-world problems such as robotic control.

HIERARCHICAL REINFORCEMENT LEARNING

Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning

NeurIPS 2018 tensorflow/models

We demonstrate this framework on 3D pose estimation by proposing a differentiable objective that seeks the optimal set of keypoints for recovering the relative pose between two views of an object.

3D POSE ESTIMATION

Searching for Efficient Multi-Scale Architectures for Dense Image Prediction

NeurIPS 2018 tensorflow/models

Recent progress has demonstrated that such meta-learning methods may exceed scalable human-invented architectures on image classification tasks.

IMAGE CLASSIFICATION META-LEARNING SEMANTIC SEGMENTATION STREET SCENE PARSING

Video-to-Video Synthesis

NeurIPS 2018 NVIDIA/vid2vid

We study the problem of video-to-video synthesis, whose goal is to learn a mapping function from an input source video (e. g., a sequence of semantic segmentation masks) to an output photorealistic video that precisely depicts the content of the source video.

SEMANTIC SEGMENTATION VIDEO PREDICTION VIDEO-TO-VIDEO SYNTHESIS

CatBoost: unbiased boosting with categorical features

NeurIPS 2018 catboost/catboost

This paper presents the key algorithmic techniques behind CatBoost, a new gradient boosting toolkit.

DIMENSIONALITY REDUCTION REGRESSION

Bilinear Attention Networks

NeurIPS 2018 facebookresearch/pythia

In this paper, we propose bilinear attention networks (BAN) that find bilinear attention distributions to utilize given vision-language information seamlessly.

VISUAL QUESTION ANSWERING

Neural Ordinary Differential Equations

NeurIPS 2018 rtqichen/torchdiffeq

Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network.

MULTIVARIATE TIME SERIES FORECASTING MULTIVARIATE TIME SERIES IMPUTATION

Glow: Generative Flow with Invertible 1x1 Convolutions

NeurIPS 2018 openai/glow

Flow-based generative models (Dinh et al., 2014) are conceptually attractive due to tractability of the exact log-likelihood, tractability of exact latent-variable inference, and parallelizability of both training and synthesis.

IMAGE GENERATION