This post teaches you how to implement your own spam filter in under 100 lines of Python code. While doing this hands-on exercise, you'll work with natural language data, learn how to detect the words spammers use automatically, and learn how to use a Naive Bayes classifier for binary classification. The task is to distinguish between two types of emails, "spam" and "non-spam" often called "ham". The machine learning classifier will detect that an email is spam if it is characterised by certain features. The textual content of the email – words like "Viagra" or "lottery" or phrases like "You've won a 100,000,000 dollars!

We show that a generative random field model, which we call generative ConvNet, can be derived from the commonly used discriminative ConvNet, by assuming a ConvNet for multi-category classification and assuming one of the categories is a base category generated by a reference distribution. If we further assume that the non-linearity in the ConvNet is Rectified Linear Unit (ReLU) and the reference distribution is Gaussian white noise, then we obtain a generative ConvNet model that is unique among energy-based models: The model is piecewise Gaussian, and the means of the Gaussian pieces are defined by an auto-encoder, where the filters in the bottom-up encoding become the basis functions in the top-down decoding, and the binary activation variables detected by the filters in the bottom-up convolution process become the coefficients of the basis functions in the top-down deconvolution process. The Langevin dynamics for sampling the generative ConvNet is driven by the reconstruction error of this auto-encoder. The contrastive divergence learning of the generative ConvNet reconstructs the training images by the auto-encoder. The maximum likelihood learning algorithm can synthesize realistic natural image patterns.

This paper introduces a new technique for quantifying the approximation error of a broad class of probabilistic inference programs, including ones based on both variational and Monte Carlo approaches. The key idea is to derive a subjective bound on the symmetrized KL divergence between the distribution achieved by an approximate inference program and its true target distribution. The bound's validity (and subjectivity) rests on the accuracy of two auxiliary probabilistic programs: (i) a "reference" inference program that defines a gold standard of accuracy and (ii) a "meta-inference" program that answers the question "what internal random choices did the original approximate inference program probably make given that it produced a particular result?" The paper includes empirical results on inference problems drawn from linear regression, Dirichlet process mixture modeling, HMMs, and Bayesian networks. The experiments show that the technique is robust to the quality of the reference inference program and that it can detect implementation bugs that are not apparent from predictive performance.

For common readers or for experts, the topic of machine learning is one that more often than not brings up lengthy heated discussions, with eyes turning and heads shaking in disagreement. No wonder why... Mounds of private information are being collected by giant corporations, stored in private data silos, and exposed to us only through creepy and yet insightful automated recommendations and suggestions. Like it or not, machine learning has entered our lives boldly and is here to stay. In the voice of Siri, in our search engines, in systems that protect us from frauds and intrusions, in applications that understand our emotions, and the list goes on and on… These days, my phone auto completes almost all information about my new contacts and meetings. I can almost feel a growing discomfort with that thought and I know I'm not alone.

Motivated by a real-world problem, we study a novel budgeted optimization problem where the goal is to optimize an unknown function f(.) given a budget by requesting a sequence of samples from the function. In our setting, however, evaluating the function at precisely specified points is not practically possible due to prohibitive costs. Instead, we can only request constrained experiments. A constrained experiment, denoted by Q, specifies a subset of the input space for the experimenter to sample the function from. The outcome of Q includes a sampled experiment x, and its function output f(x). Importantly, as the constraints of Q become looser, the cost of fulfilling the request decreases, but the uncertainty about the location x increases. Our goal is to manage this trade-off by selecting a set of constrained experiments that best optimize f(.) within the budget. We study this problem in two different settings, the non-sequential (or batch) setting where a set of constrained experiments is selected at once, and the sequential setting where experiments are selected one at a time. We evaluate our proposed methods for both settings using synthetic and real functions. The experimental results demonstrate the efficacy of the proposed methods.

This paper proposes CF-NADE, a neural autoregressive architecture for collaborative filtering (CF) tasks, which is inspired by the Restricted Boltzmann Machine (RBM) based CF model and the Neural Autoregressive Distribution Estimator (NADE). We first describe the basic CF-NADE model for CF tasks. Then we propose to improve the model by sharing parameters between different ratings. A factored version of CF-NADE is also proposed for better scalability. Furthermore, we take the ordinal nature of the preferences into consideration and propose an ordinal cost to optimize CF-NADE, which shows superior performance. Finally, CF-NADE can be extended to a deep model, with only moderately increased computational complexity. Experimental results show that CF-NADE with a single hidden layer beats all previous state-of-the-art methods on MovieLens 1M, MovieLens 10M, and Netflix datasets, and adding more hidden layers can further improve the performance.

Black box variational inference allows researchers to easily prototype and evaluate an array of models. Recent advances allow such algorithms to scale to high dimensions. However, a central question remains: How to specify an expressive variational distribution that maintains efficient computation? To address this, we develop hierarchical variational models (HVMs). HVMs augment a variational approximation with a prior on its parameters, which allows it to capture complex structure for both discrete and continuous latent variables. The algorithm we develop is black box, can be used for any HVM, and has the same computational efficiency as the original approximation. We study HVMs on a variety of deep discrete latent variable models. HVMs generalize other expressive variational distributions and maintains higher fidelity to the posterior.

We propose a new reinforcement learning algorithm for partially observable Markov decision processes (POMDP) based on spectral decomposition methods. While spectral methods have been previously employed for consistent learning of (passive) latent variable models such as hidden Markov models, POMDPs are more challenging since the learner interacts with the environment and possibly changes the future observations in the process. We devise a learning algorithm running through episodes, in each episode we employ spectral techniques to learn the POMDP parameters from a trajectory generated by a fixed policy. At the end of the episode, an optimization oracle returns the optimal memoryless planning policy which maximizes the expected reward based on the estimated POMDP model. We prove an order-optimal regret bound with respect to the optimal memoryless policy and efficient scaling with respect to the dimensionality of observation and action spaces.

We present a new approach to estimating the interdependence of industries in an economy by applying data science solutions. By exploiting interfirm buyer--seller network data, we show that the problem of estimating the interdependence of industries is similar to the problem of uncovering the latent block structure in network science literature. To estimate the underlying structure with greater accuracy, we propose an extension of the sparse block model that incorporates node textual information and an unbounded number of industries and interactions among them. The latter task is accomplished by extending the well-known Chinese restaurant process to two dimensions. Inference is based on collapsed Gibbs sampling, and the model is evaluated on both synthetic and real-world datasets. We show that the proposed model improves in predictive accuracy and successfully provides a satisfactory solution to the motivated problem. We also discuss issues that affect the future performance of this approach.

A pdf version of this document created using latex can be downloaded by clicking here. Polymorphic malware detection is challenging due to the continual mutations miscreants introduce to successive instances of a particular virus. Such changes are akin to mutations in biological sequences. Recently, high-throughput methods for gene sequence classification have been developed by the bioinformatics and computational biology communities. In this paper, we argue that these methods can be usefully applied to malware detection. Unfortunately, gene classification tools are usually optimized for and restricted to an alphabet of four letters (nucleic acids). Consequently, we have selected the Strand gene sequence classifier, which offers a robust classification strategy that can easily accommodate unstructured data with any alphabet including source code or compiled machine code. To demonstrate Stand's suitability for classifying malware, we execute it on approximately 500GB of malware data provided by the Kaggle Microsoft Malware Classification Challenge (BIG 2015) used for predicting 9 classes of polymorphic malware.