Microsoft Research Blog

Artificial intelligence

  1. Deep High-Resolution Representation Learning for Human Pose Estimation 

    February 24, 2019 | Ke Sun, Bin Xiao, Dong Liu, and Jingdong Wang

    In this paper, we are interested in the human pose estimation problem with a focus on learning reliable high-resolution representations. Most existing methods recover high-resolution representations from low-resolution representations produced by a high-to-low resolution network. Instead, our proposed network maintains high-resolution representations through the whole…

  3. NAIL: A General Interactive Fiction Agent 

    February 14, 2019

    Interactive Fiction (IF) games are complex textual decision making problems. This paper introduces NAIL, an autonomous agent for general parser-based IF games. NAIL won the 2018 Text Adventure AI Competition, where it was evaluated on twenty unseen games. This paper describes the architecture, development, and…

  4. Graph-Based Skill Acquisition For Reinforcement Learning 

    February 12, 2019 | Matheus R. F. Mendonça, Artur Ziviani, and André da Motta Salles Barreto

    In machine learning, Reinforcement Learning (RL) is an important tool for creating intelligent agents that learn solely through experience. One particular subarea within the RL domain that has received great attention is how to define macro-actions, which are temporal abstractions composed of a sequence of…

  5. Planning with actively eliciting preferences 

    February 1, 2019

    Planning with preferences has been employed extensively to quickly generate high-quality plans. However, it may be difficult for the human expert to supply this information without knowledge of the reasoning employed by the planner. We consider the problem of actively eliciting preferences from a human…

  6. A Distillation Approach to Data Efficient Individual Treatment Effect Estimation 

    January 27, 2019 | Maggie Makar, Adith Swaminathan, and Emre Kiciman

    The potential for using machine learning algorithms as a tool for suggesting optimal interventions has fueled significant interest in developing methods for estimating heterogeneous or individual treatment effects (ITEs) from observational data. While several methods for estimating ITEs have been recently suggested, these methods assume…

  7. Interpreting Deep Neural Networks Through Variable Importance. 

    January 27, 2019

    While the success of deep neural networks (DNNs) is well-established across a variety of domains, our ability to explain and interpret these methods is limited. Unlike previously proposed local methods which try to explain particular classification decisions, we focus on global interpretability and ask a…

  8. Uncovering and Mitigating Algorithmic Bias through Learned Latent Structure 

    January 26, 2019

    Recent research has highlighted the vulnerabilities of modern machine learning based systems to bias, especially towards segments of society that are under-represented in training data. In this work, we develop a novel, tunable algorithm for mitigating the hidden, and potentially unknown, biases within training data.…

  9. CariGANs: unpaired photo-to-caricature translation 

    January 9, 2019 | Kaidi Cao, Jing Liao, and Lu Yuan

    Facial caricature is an art form of drawing faces in an exaggerated way to convey humor or sarcasm. In this paper, we propose the first Generative Adversarial Network (GAN) for unpaired photo-to-caricature translation, which we call "CariGANs". It explicitly models geometric exaggeration and appearance stylization…

  11. Entropy and mutual information in models of deep neural networks 

    December 31, 2018

    We examine a class of stochastic deep learning models with a tractable method to compute information-theoretic quantities. Our contributions are three-fold: (i) We show how entropies and mutual informations can be derived from heuristic statistical physics methods, under the assumption that weight matrices are independent…

  12. Gated Context Aggregation Network for Image Dehazing and Deraining 

    December 31, 2018

    Image dehazing aims to recover the uncorrupted content from a hazy image. Instead of leveraging traditional low-level or handcrafted image priors as the restoration constraints, e.g., dark channels and increased contrast, we propose an end-to-end gated context aggregation network to directly restore the final haze-free…