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

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

GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration

NeurIPS 2018 cornellius-gp/gpytorch

Despite advances in scalable models, the inference tools used for Gaussian processes (GPs) have yet to fully capitalize on developments in computing hardware.

Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents

NeurIPS 2018 uber-common/deep-neuroevolution

Evolution strategies (ES) are a family of black-box optimization algorithms able to train deep neural networks roughly as well as Q-learning and policy gradient methods on challenging deep reinforcement learning (RL) problems, but are much faster (e. g. hours vs. days) because they parallelize better.

POLICY GRADIENT METHODS Q-LEARNING

GILBO: One Metric to Measure Them All

NeurIPS 2018 google/compare_gan

We propose a simple, tractable lower bound on the mutual information contained in the joint generative density of any latent variable generative model: the GILBO (Generative Information Lower BOund).

Visual Reinforcement Learning with Imagined Goals

NeurIPS 2018 vitchyr/rlkit

For an autonomous agent to fulfill a wide range of user-specified goals at test time, it must be able to learn broadly applicable and general-purpose skill repertoires.

UNSUPERVISED REPRESENTATION LEARNING

PointCNN: Convolution On X-Transformed Points

NeurIPS 2018 yangyanli/PointCNN

We present a simple and general framework for feature learning from point cloud.

Multiple Instance Learning for Efficient Sequential Data Classification on Resource-constrained Devices

NeurIPS 2018 Microsoft/EdgeML

We propose a method, EMI-RNN, that exploits these observations by using a multiple instance learning formulation along with an early prediction technique to learn a model that achieves better accuracy compared to baseline models, while simultaneously reducing computation by a large fraction.

MULTIPLE INSTANCE LEARNING TIME SERIES TIME SERIES CLASSIFICATION

FastGRNN: A Fast, Accurate, Stable and Tiny Kilobyte Sized Gated Recurrent Neural Network

NeurIPS 2018 Microsoft/EdgeML

FastRNN addresses these limitations by adding a residual connection that does not constrain the range of the singular values explicitly and has only two extra scalar parameters.

ACTION CLASSIFICATION LANGUAGE MODELLING SPEECH RECOGNITION TIME SERIES TIME SERIES CLASSIFICATION

Deep Neural Networks with Box Convolutions

NeurIPS 2018 shrubb/box-convolutions

Box filters computed using integral images have been part of the computer vision toolset for a long time.

SEMANTIC SEGMENTATION