Neural Architecture Search (NAS) has seen an explosion of research in the past few years. A variety of methods have been proposed to perform NAS, including reinforcement learning, Bayesian optimization with a Gaussian process model, evolutionary search, and gradient descent. In this work, we design a NAS algorithm that performs Bayesian optimization using a neural network model. We develop a path-based encoding scheme to featurize the neural architectures that are used to train the neural network model. This strategy is particularly effective for encoding architectures in cell-based search spaces. After training on just 200 random neural architectures, we are able to predict the validation accuracy of a new architecture to within one percent of its true accuracy on average, for popular search spaces. This may be of independent interest beyond Bayesian neural architecture search. We test our algorithm on the NASBench (Ying et al. 2019) and DARTS (Liu et al. 2018) search spaces, and we show that our algorithm outperforms other NAS methods including evolutionary search, reinforcement learning, AlphaX, ASHA, and DARTS. Our algorithm is over 100x more efficient than random search, and 3.8x more efficient than the next-best algorithm on the NASBench dataset. As there have been problems with fair and reproducible experimental evauations in the field of NAS, we adhere to the recent NAS research checklist (Lindauer and Hutter 2019) to facilitate NAS research. In particular, our implementation has been made publicly available, including all details needed to fully reproduce our results.

Non-intrusive load monitoring (NILM) aims at separating a whole-home energy signal into its appliance components. Such method can be harnessed to provide various services to better manage and control energy consumption (optimal planning and saving). NILM has been traditionally approached from signal processing and electrical engineering perspectives. Recently, machine learning has started to play an important role in NILM. While most work has focused on supervised algorithms, unsupervised approaches can be more interesting and of practical use in real case scenarios. Specifically, they do not require labelled training data to be acquired from individual appliances and the algorithm can be deployed to operate on the measured aggregate data directly. In this paper, we propose a fully unsupervised NILM framework based on Bayesian hierarchical mixture models. In particular, we develop a new method based on Gaussian Latent Dirichlet Allocation (GLDA) in order to extract global components that summarise the energy signal. These components provide a representation of the consumption patterns. Designed to cope with big data, our algorithm, unlike existing NILM ones, does not focus on appliance recognition. To handle this massive data, GLDA works online. Another novelty of this work compared to the existing NILM is that the data involves different utilities (e.g, electricity, water and gas) as well as some sensors measurements. Finally, we propose different evaluation methods to analyse the results which show that our algorithm finds useful patterns.

This paper introduces a new notion of dimensionality of probabilistic models from an information-theoretic view point. We call it the "descriptive dimension"(Ddim). We show that Ddim coincides with the number of independent parameters for the parametric class, and can further be extended to real-valued dimensionality when a number of models are mixed. The paper then derives the rate of convergence of the MDL (Minimum Description Length) learning algorithm which outputs a normalized maximum likelihood (NML) distribution with model of the shortest NML codelength. The paper proves that the rate is governed by Ddim. The paper also derives error probabilities of the MDL-based test for multiple model change detection. It proves that they are also governed by Ddim. Through the analysis, we demonstrate that Ddim is an intrinsic quantity which characterizes the performance of the MDL-based learning and change detection.

In recent years, the interest in leveraging quantum effects for enhancing machine learning tasks has significantly increased. Many algorithms speeding up supervised and unsupervised learning were established. The first framework in which ways to exploit quantum resources specifically for the broader context of reinforcement learning were found is projective simulation. Projective simulation presents an agent-based reinforcement learning approach designed in a manner which may support quantum walk-based speed-ups. Although classical variants of projective simulation have been benchmarked against common reinforcement learning algorithms, very few formal theoretical analyses have been provided for its performance in standard learning scenarios. In this paper, we provide a detailed formal discussion of the properties of this model. Specifically, we prove that one version of the projective simulation model, understood as a reinforcement learning approach, converges to optimal behavior in a large class of Markov decision processes. This proof shows that a physically-inspired approach to reinforcement learning can guarantee to converge.

Robust MDPs are a promising framework for computing robust policies in reinforcement learning. Ambiguity sets, which represent the plausible errors in transition probabilities, determine the trade-off between robustness and average-case performance. The standard practice of defining ambiguity sets using the $L_1$ norm leads, unfortunately, to loose and impractical guarantees. This paper describes new methods for optimizing the shape of ambiguity sets beyond the $L_1$ norm. We derive new high-confidence sampling bounds for weighted $L_1$ and weighted $L_\infty$ ambiguity sets and describe how to compute near-optimal weights from rough value function estimates. Experimental results on a diverse set of benchmarks show that optimized ambiguity sets provide significantly tighter robustness guarantees.

Let's step back first and frame our classification problem in Bayesian terms -- where we have a set of prior beliefs and update our beliefs as we observe and collect evidence. In statistics, everything revolves around hypotheses. We make a hypothesis (an informed guess) about how the world works, and then we go about collecting evidence to test that hypothesis (if you would like to know the details, I wrote a post about hypothesis testing here). Classification models can be framed as a hypothesis as well. Let's first write out the objective and variables of our classification problem: OK, so that's classification -- now let's examine classification through a Bayesian lens.

Estimating machine learning performance 'in the wild' is both an important and unsolved problem. In this paper, we seek to examine, understand, and predict the pointwise competence of classification models. Our contributions are twofold: First, we establish a statistically rigorous definition of competence that generalizes the common notion of classifier confidence; second, we present the ALICE (Accurate Layerwise Interpretable Competence Estimation) Score, a pointwise competence estimator for any classifier. By considering distributional, data, and model uncertainty, ALICE empirically shows accurate competence estimation in common failure situations such as class-imbalanced datasets, out-of-distribution datasets, and poorly trained models. Our contributions allow us to accurately predict the competence of any classification model given any input and error function. We compare our score with state-of-the-art confidence estimators such as model confidence and Trust Score, and show significant improvements in competence prediction over these methods on datasets such as DIGITS, CIFAR10, and CIFAR100.

We consider a novel application of inverse reinforcement learning which involves modeling, learning and predicting the commenting behavior of YouTube viewers. Each group of users is modeled as a rationally inattentive Bayesian agent. Our methodology integrates three key components. First, to identify distinct commenting patterns, we use deep embedded clustering to estimate framing information (essential extrinsic features) that clusters users into distinct groups. Second, we present an inverse reinforcement learning algorithm that uses Bayesian revealed preferences to test for rationality: does there exist a utility function that rationalizes the given data, and if yes, can it be used to predict future behavior? Finally, we impose behavioral economics constraints stemming from rational inattention to characterize the attention span of groups of users.The test imposes a R{\'e}nyi mutual information cost constraint which impacts how the agent can select attention strategies to maximize their expected utility. After a careful analysis of a massive YouTube dataset, our surprising result is that in most YouTube user groups, the commenting behavior is consistent with optimizing a Bayesian utility with rationally inattentive constraints. The paper also highlights how the rational inattention model can accurately predict future commenting behavior. The massive YouTube dataset and analysis used in this paper are available on GitHub and completely reproducible.

Causal knowledge is vital for effective reasoning in science, as causal relations, unlike correlations, allow one to reason about the outcomes of interventions. Algorithms that can discover causal relations from observational data are based on the assumption that all variables have been jointly measured in a single dataset. In many cases this assumption fails. Previous approaches to overcoming this shortcoming devised algorithms that returned all joint causal structures consistent with the conditional independence information contained in each individual dataset. But, as conditional independence tests only determine causal structure up to Markov equivalence, the number of consistent joint structures returned by these approaches can be quite large. The last decade has seen the development of elegant algorithms for discovering causal relations beyond conditional independence, which can distinguish among Markov equivalent structures. In this work we adapt and extend these so-called bivariate causal discovery algorithms to the problem of learning consistent causal structures from multiple datasets with overlapping variables belonging to the same generating process, providing a sound and complete algorithm that outperforms previous approaches on synthetic and real data.

We begin by reiterating that common neural network activation functions have simple Bayesian origins. In this spirit, we go on to show that Bayes's theorem also implies a simple recurrence relation; this leads to a Bayesian recurrent unit with a prescribed feedback formulation. We show that introduction of a context indicator leads to a variable feedback that is similar to the forget mechanism in conventional recurrent units. A similar approach leads to a probabilistic input gate. The Bayesian formulation leads naturally to the two pass algorithm of the Kalman smoother or forward-backward algorithm, meaning that inference naturally depends upon future inputs as well as past ones. Experiments on speech recognition confirm that the resulting architecture can perform as well as a bidirectional recurrent network with the same number of parameters as a unidirectional one. Further, when configured explicitly bidirectionally, the architecture can exceed the performance of a conventional bidirectional recurrence.