This dissertation investigates the use of one-sided classification algorithms in the application of separating hazardous chlorinated solvents from other materials, based on their Raman spectra. The experimentation is carried out using a new one-sided classification toolkit that was designed and developed from the ground up. In the one-sided classification paradigm, the objective is to separate elements of the target class from all outliers. These one-sided classifiers are generally chosen, in practice, when there is a deficiency of some sort in the training examples. Sometimes outlier examples can be rare, expensive to label, or even entirely absent. However, this author would like to note that they can be equally applicable when outlier examples are plentiful but nonetheless not statistically representative of the complete outlier concept. It is this scenario that is explicitly dealt with in this research work. In these circumstances, one-sided classifiers have been found to be more robust that conventional multi-class classifiers. The term "unexpected" outliers is introduced to represent outlier examples, encountered in the test set, that have been taken from a different distribution to the training set examples. These are examples that are a result of an inadequate representation of all possible outliers in the training set. It can often be impossible to fully characterise outlier examples given the fact that they can represent the immeasurable quantity of "everything else" that is not a target. The findings from this research have shown the potential drawbacks of using conventional multi-class classification algorithms when the test data come from a completely different distribution to that of the training samples.

For many modern applications in science and engineering, data are collected in a streaming fashion carrying time-varying information, and practitioners need to process them with a limited amount of memory and computational resources in a timely manner for decision making. This often is coupled with the missing data problem, such that only a small fraction of data attributes are observed. These complications impose significant, and unconventional, constraints on the problem of streaming Principal Component Analysis (PCA) and subspace tracking, which is an essential building block for many inference tasks in signal processing and machine learning. This survey article reviews a variety of classical and recent algorithms for solving this problem with low computational and memory complexities, particularly those applicable in the big data regime with missing data. We illustrate that streaming PCA and subspace tracking algorithms can be understood through algebraic and geometric perspectives, and they need to be adjusted carefully to handle missing data. Both asymptotic and non-asymptotic convergence guarantees are reviewed. Finally, we benchmark the performance of several competitive algorithms in the presence of missing data for both well-conditioned and ill-conditioned systems.

Sample efficiency is critical in solving real-world reinforcement learning problems, where agent-environment interactions can be costly. Imitation learning from expert advice has proved to be an effective strategy for reducing the number of interactions required to train a policy. Online imitation learning, a specific type of imitation learning that interleaves policy evaluation and policy optimization, is a particularly effective framework for training policies with provable performance guarantees. In this work, we seek to further accelerate the convergence rate of online imitation learning, making it more sample efficient. We propose two model-based algorithms inspired by Follow-the-Leader (FTL) with prediction: MoBIL-VI based on solving variational inequalities and MoBIL-Prox based on stochastic first-order updates. When a dynamics model is learned online, these algorithms can provably accelerate the best known convergence rate up to an order. Our algorithms can be viewed as a generalization of stochastic Mirror-Prox by Juditsky et al. (2011), and admit a simple constructive FTL-style analysis of performance. The algorithms are also empirically validated in simulation.

Predictive models that are developed in a regulated industry or a regulated application, like determination of credit worthiness, must be interpretable and rational (e.g., meaningful improvements in basic credit behavior must result in improved credit worthiness scores). Machine Learning technologies provide very good performance with minimal analyst intervention, making them well suited to a high volume analytic environment, but the majority are black box tools that provide very limited insight or interpretability into key drivers of model performance or predicted model output values. This paper presents a methodology that blends one of the most popular predictive statistical modeling methods for binary classification with a core model enhancement strategy found in machine learning. The resulting prediction methodology provides solid performance, from minimal analyst effort, while providing the interpretability and rationality required in regulated industries, as well as in other environments where interpretation of model parameters is required (e.g. businesses that require interpretation of models, to take action on them).

There is growing evidence that converting targets to soft targets in supervised learning can provide considerable gains in performance. Much of this work has considered classification, converting hard zero-one values to soft labels---such as by adding label noise, incorporating label ambiguity or using distillation. In parallel, there is some evidence from a regression setting in reinforcement learning that learning distributions can improve performance. In this work, we investigate the reasons for this improvement, in a regression setting. We introduce a novel distributional regression loss, and similarly find it significantly improves prediction accuracy. We investigate several common hypotheses, around reducing overfitting and improved representations. We instead find evidence for an alternative hypothesis: this loss is easier to optimize, with better behaved gradients, resulting in improved generalization. We provide theoretical support for this alternative hypothesis, by characterizing the norm of the gradients of this loss.

Timely assessment of compound toxicity is one of the biggest challenges facing the pharmaceutical industry today. A significant proportion of compounds identified as potential leads are ultimately discarded due to the toxicity they induce. In this paper, we propose a novel machine learning approach for the prediction of molecular activity on ToxCast targets. We combine extreme gradient boosting with fully-connected and graph-convolutional neural network architectures trained on QSAR physical molecular property descriptors, PubChem molecular fingerprints, and SMILES sequences. Our ensemble predictor leverages the strengths of each individual technique, significantly outperforming existing state-of-the art models on the ToxCast and Tox21 toxicity-prediction datasets. We provide free access to molecule toxicity prediction using our model at http://www.owkin.com/toxicblend.

Complex computer simulations are commonly required for accurate data modelling in many scientific disciplines, making statistical inference challenging due to the intractability of the likelihood evaluation for the observed data. Furthermore, sometimes one is interested on inference drawn over a subset of the generative model parameters while taking into account model uncertainty or misspecification on the remaining nuisance parameters. In this work, we show how non-linear summary statistics can be constructed by minimising inference-motivated losses via stochastic gradient descent.

Generative models with both discrete and continuous latent variables are highly motivated by the structure of many real-world data sets. They present, however, subtleties in training often manifesting in the discrete latent being under leveraged. In this paper, we show that such models are more amenable to training when using the Optimal Transport framework of Wasserstein Autoencoders. We find our discrete latent variable to be fully leveraged by the model when trained, without any modifications to the objective function or significant fine tuning. Our model generates comparable samples to other approaches while using relatively simple neural networks, since the discrete latent variable carries much of the descriptive burden. Furthermore, the discrete latent provides significant control over generation.

We consider the binary classification problem in which the objective function is the exponential loss with a ReLU model, and study the convergence property of the stochastic gradient descent (SGD) algorithm on linearly separable data. We show that the gradient descent (GD) algorithm do not always learn desirable model parameters due to the nonlinear ReLU model. Then, we identify a certain condition of data samples, under which we show that SGD can learn a proper classifier with implicit bias. In specific, we establish the sub-linear convergence rate of the function value generated by SGD to global minimum. We further show that SGD actually converges in expectation to the maximum margin classifier with respect to the samples with +1 label under the ReLU model at the rate O(1/ln t). We also extend our study to the case of multi-ReLU neurons, and show that SGD converges to a certain non-linear maximum margin classifier for a class of non-linearly separable data.

Stochastic gradient Markov chain Monte Carlo (SG-MCMC) has become increasingly popular for simulating posterior samples in large-scale Bayesian modeling. However, existing SG-MCMC schemes are not tailored to any specific probabilistic model, even a simple modification of the underlying dynamical system requires significant physical intuition. This paper presents the first meta-learning algorithm that allows automated design for the underlying continuous dynamics of an SG-MCMC sampler. The learned sampler generalizes Hamiltonian dynamics with state-dependent drift and diffusion, enabling fast traversal and efficient exploration of neural network energy landscapes. Experiments validate the proposed approach on both Bayesian fully connected neural network and Bayesian recurrent neural network tasks, showing that the learned sampler out-performs generic, hand-designed SG-MCMC algorithms, and generalizes to different datasets and larger architectures.