In real-world applications, it is conceivable that only positive and unlabeled (PU) data are available for training a classifier. For instance, in land-cover image classification, images of urban regions can be easily labeled, while images of non-urban regions are difficult to annotate due to high diversity of non-urban regions containing, e.g., forest, seas, grasses, and soil (Li et al., 2011). To cope with such situations, PU classification has been actively studied (Letouzey et al., 2000; Elkan and Noto, 2008; du Plessis et al., 2015), and the state-of-the-art method allows us to systematically train deep neural networks only from PU data (Kiryo et al., 2017). However, existing PU classification methods typically require an estimate of the class-prior probability, and their performance is sensitive to the quality of class-prior estimation (Kiryo et al., 2017). Although various class-prior estimation methods from PU data have been proposed so far (du Plessis and Sugiyama, 2014; Ramaswamy et al., 2016; Jain et al., 2016; du Plessis et al., 2017; Northcutt et al., 2017), accurate estimation of the class-prior is still highly challenging particularly for high-dimensional data.

Interpretable machine learning models have received increasing interest in recent years, especially in domains where humans are involved in the decision-making process. However, the possible loss of the task performance for gaining interpretability is often inevitable. This performance downgrade puts practitioners in a dilemma of choosing between a top-performing black-box model with no explanations and an interpretable model with unsatisfying task performance. In this work, we propose a novel framework for building a Hybrid Decision Model that integrates an interpretable model with any black-box model to introduce explanations in the decision making process while preserving or possibly improving the predictive accuracy. We propose a novel metric, explainability, to measure the percentage of data that are sent to the interpretable model for decision. We also design a principled objective function that considers predictive accuracy, model interpretability, and data explainability. Under this framework, we develop Collaborative Black-box and RUle Set Hybrid (CoBRUSH) model that combines logic rules and any black-box model into a joint decision model. An input instance is first sent to the rules for decision. If a rule is satisfied, a decision will be directly generated. Otherwise, the black-box model is activated to decide on the instance. To train a hybrid model, we design an efficient search algorithm that exploits theoretically grounded strategies to reduce computation. Experiments show that CoBRUSH models are able to achieve same or better accuracy than their black-box collaborator working alone while gaining explainability. They also have smaller model complexity than interpretable baselines.

Modern reinforcement learning algorithms reach super-human performance in many board and video games, but they are sample inefficient, i.e. they typically require significantly more playing experience than humans to reach an equal performance level. To improve sample efficiency, an agent may build a model of the environment and use planning methods to update its policy. In this article we introduce VaST (Variational State Tabulation), which maps an environment with a high-dimensional state space (e.g. the space of visual inputs) to an abstract tabular environment. Prioritized sweeping with small backups, a highly efficient planning method, can then be used to update state-action values. We show how VaST can rapidly learn to maximize reward in tasks like 3D navigation and efficiently adapt to sudden changes in rewards or transition probabilities.

We provide a numerically robust and fast method capable of exploiting the local geometry when solving large-scale stochastic optimisation problems. Our key innovation is an auxiliary variable construction coupled with an inverse Hessian approximation computed using a receding history of iterates and gradients. It is the Markov chain nature of the classic stochastic gradient algorithm that enables this development. The construction offers a mechanism for stochastic line search adapting the step length. We numerically evaluate and compare against current state-of-the-art with encouraging performance on real-world benchmark problems where the number of observations and unknowns is in the order of millions.

Markov processes, both classical and higher order, are often used to model dynamic processes, such as stock prices, molecular dynamics, and Monte Carlo methods. Previous works have shown that an autoencoder can be formulated as a specific type of Markov chain. Here, we propose a generative neural network known as a transition encoder, or transcoder, which learns such continuous-state dynamic processes. We show that the transcoder is able to learn both deterministic and stochastic dynamic processes on several systems. We explore a number of applications of the transcoder including generating unseen trajectories and examining the propensity for chaos in a dynamic system. Further, we show that the transcoder can speed up Markov Chain Monte Carlo (MCMC) sampling to a convergent distribution by training it to make several steps at a time. Finally, we show that the hidden layers of a transcoder are useful for visualization and salient feature extraction of the transition process itself.

We propose a novel class of time-varying nonparanormal graphical models, which allows us to model high dimensional heavy-tailed systems and the evolution of their latent network structures. Under this model, we develop statistical tests for presence of edges both locally at a fixed index value and globally over a range of values. The tests are developed for a high-dimensional regime, are robust to model selection mistakes and do not require commonly assumed minimum signal strength. The testing procedures are based on a high dimensional, debiasing-free moment estimator, which uses a novel kernel smoothed Kendall's tau correlation matrix as an input statistic. The estimator consistently estimates the latent inverse Pearson correlation matrix uniformly in both the index variable and kernel bandwidth. Its rate of convergence is shown to be minimax optimal. Our method is supported by thorough numerical simulations and an application to a neural imaging data set.

A key phase in the bridge design process is the selection of the structural system. Due to budget and time constraints, engineers typically rely on engineering judgment and prior experience when selecting a structural system, often considering a limited range of design alternatives. The objective of this study was to explore the suitability of supervised machine learning as a preliminary design aid that provides guidance to engineers with regards to the statistically optimal bridge type to choose, ultimately improving the likelihood of optimized design, design standardization, and reduced maintenance costs. In order to devise this supervised learning system, data for over 600,000 bridges from the National Bridge Inventory database were analyzed. Key attributes for determining the bridge structure type were identified through three feature selection techniques. Potentially useful attributes like seismic intensity and historic data on the cost of materials (steel and concrete) were then added from the US Geological Survey (USGS) database and Engineering News Record. Decision tree, Bayes network and Support Vector Machines were used for predicting the bridge design type. Due to state-to-state variations in material availability, material costs, and design codes, supervised learning models based on the complete data set did not yield favorable results. Supervised learning models were then trained and tested using 10-fold cross validation on data for each state. Inclusion of seismic data improved the model performance noticeably. The data was then resampled to reduce the bias of the models towards more common design types, and the supervised learning models thus constructed showed further improvements in performance. The average recall and precision for the state models was 88.6% and 88.0% using Decision Trees, 84.0% and 83.7% using Bayesian Networks, and 80.8% and 75.6% using SVM.

Learn how to to scale up a machine learning algorithm. The cloud and machine learning: two phrases with a lot of hype that few people understand. We're intimately familiar with both here at Leverege, so hopefully this article will shed some light on the two topics. Before we share what we've learned using Google Cloud ML Engine, we need to do a quick refresher on how machine learning is done in production. So how does Google Cloud ML fit into all of this?

The academic Deep Learning research community has largely stayed away from the financial markets. Maybe that's because the finance industry has a bad reputation, the problem doesn't seem interesting from a research perspective, or because data is difficult and expensive to obtain. In this post, I'm going to argue that training Reinforcement Learning agents to trade in the financial (and cryptocurrency) markets can be an extremely interesting research problem. I believe that it has not received enough attention from the research community but has the potential to push the state-of-the art of many related fields. It is quite similar to training agents for multiplayer games such as DotA, and many of the same research problems carry over. Knowing virtually nothing about trading, I have spent the past few months working on a project in this field. This is not a "price prediction using Deep Learning" post. So, if you're looking for example code and models you may be disappointed. Instead, I want to talk on a more high level about why learning to trade using Machine Learning is difficult, what some of the challenges are, and where I think Reinforcement Learning fits in. If there's enough interest in this area I may follow up with another post that includes concrete examples. I expect most readers to have no background in trading, just like I didn't, so I will start out with covering some of the basics. I'm by no means an expert, so please let me know in the comments so if you find mistakes. I will use cryptocurrencies as a running example in this post, but the same concepts apply to most of the financial markets. The reason to use cryptocurrencies is that data is free, public, and easily accessible. Anyone can sign up to trade. The barriers to trading in the financial markets are a little higher, and data can be expensive.

In spoken dialogue systems, we aim to deploy artificial intelligence to build automated dialogue agents that can converse with humans. A part of this effort is the policy optimisation task, which attempts to find a policy describing how to respond to humans, in the form of a function taking the current state of the dialogue and returning the response of the system. In this paper, we investigate deep reinforcement learning approaches to solve this problem. Particular attention is given to actor-critic methods, off-policy reinforcement learning with experience replay, and various methods aimed at reducing the bias and variance of estimators. When combined, these methods result in the previously proposed ACER algorithm that gave competitive results in gaming environments. These environments however are fully observable and have a relatively small action set so in this paper we examine the application of ACER to dialogue policy optimisation. We show that this method beats the current state-of-the-art in deep learning approaches for spoken dialogue systems. This not only leads to a more sample efficient algorithm that can train faster, but also allows us to apply the algorithm in more difficult environments than before. We thus experiment with learning in a very large action space, which has two orders of magnitude more actions than previously considered. We find that ACER trains significantly faster than the current state-of-the-art.