Numerous lung diseases, such as idiopathic pulmonary fibrosis (IPF), exhibit dilation of the airways. Accurate measurement of dilatation enables assessment of the progression of disease. Unfortunately the combination of image noise and airway bifurcations causes high variability in the profiles of cross-sectional areas, rendering the identification of affected regions very difficult. Here we introduce a noise-robust method for automatically detecting the location of progressive airway dilatation given two profiles of the same airway acquired at different time points. We propose a probabilistic model of abrupt relative variations between profiles and perform inference via Reversible Jump Markov Chain Monte Carlo sampling. We demonstrate the efficacy of the proposed method on two datasets; (i) images of healthy airways with simulated dilatation; (ii) pairs of real images of IPF-affected airways acquired at 1 year intervals. Our model is able to detect the starting location of airway dilatation with an accuracy of 2.5mm on simulated data. The experiments on the IPF dataset display reasonable agreement with radiologists. We can compute a relative change in airway volume that may be useful for quantifying IPF disease progression.

Given a labeled dataset that contains a rare (or minority) class of of-interest instances, as well as a large class of instances that are not of interest, how can we learn to recognize future of-interest instances over a continuous stream? We introduce RaRecognize, which (i) estimates a general decision boundary between the rare and the majority class, (ii) learns to recognize individual rare subclasses that exist within the training data, as well as (iii) flags instances from previously unseen rare subclasses as newly emerging. The learner in (i) is general in the sense that by construction it is dissimilar to the specialized learners in (ii), thus distinguishes minority from the majority without overly tuning to what is seen in the training data. Thanks to this generality, RaRecognize ignores all future instances that it labels as majority and recognizes the recurrent as well as emerging rare subclasses only. This saves effort at test time as well as ensures that the model size grows moderately over time as it only maintains specialized minority learners. Through extensive experiments, we show that RaRecognize outperforms state-of-the art baselines on three real-world datasets that contain corporate-risk and disaster documents as rare classes.

I argue that regularizing terms in standard regression methods not only help against overfitting finite data, but sometimes also yield better causal models in the infinite sample regime. I first consider a multi-dimensional variable linearly influencing a target variable with some multi-dimensional unobserved common cause, where the confounding effect can be decreased by keeping the penalizing term in Ridge and Lasso regression even in the population limit. Choosing the size of the penalizing term, is however challenging, because cross validation is pointless. Here it is done by first estimating the strength of confounding via a method proposed earlier, which yielded some reasonable results for simulated and real data. Further, I prove a `causal generalization bound' which states (subject to a particular model of confounding) that the error made by interpreting any non-linear regression as causal model can be bounded from above whenever functions are taken from a not too rich class. In other words, the bound guarantees "generalization" from observational to interventional distributions, which is usually not subject of statistical learning theory (and is only possible due to the underlying symmetries of the confounder model).

Dimensionality reduction has always been a trendy topic in machine learning. Linear subspace method for reduction, e.g., Principal Component Analysis and its variation have been widely studied[21, 16, 23], and some pieces of literature introduce probability and randomness to realize PCA[18, 12, 11, 14, 20]. However, linear subspace is not applicable when the data lies in a nonlinear manifold[2, 15]. Due to the direct connection with PCA, Singular Value Decomposition (SVD) is one of the most well-known algorithms for low-rank approximation[19, 9], and it has been widely used throughout the machine learning and statistics community. Some implementations of SVD are solving least squares[10, 22], latent semantic analysis[7, 13], genetic analysis, matrix completion[17, 3, 5, 4], data mining[6, 1] etc. However, when it comes to a large scale matrix, the runtime of traditional SVD is intolerable and the memory usage could be enormously consuming. Notations: we have a matrix A with size m n, usually m n. Our goal is to find the Principal Components (PCs) given the cumulative explained variance threshold t.

Spike sorting is a fundamental preprocessing step in neuroscience that is central to access simultaneous but distinct neuronal activities and therefore to better understand the animal or even human brain. But numerical complexity limits studies that require processing large scale datasets in terms of number of electrodes, neurons, spikes and length of the recorded signals. We propose in this work a novel active set algorithm aimed at solving the Lasso for a classical convolutional model. Our algorithm can be implemented efficiently on parallel architecture and has a linear complexity w.r.t. the temporal dimensionality which ensures scaling and will open the door to online spike sorting. We provide theoretical results about the complexity of the algorithm and illustrate it in numerical experiments along with results about the accuracy of the spike recovery and robustness to the regularization parameter.

In this article we investigate the outcomes of the standard Least Angle Regression (LAR) algorithm in high dimensions under the Gaussian noise assumption. We give the exact law of the sequence of knots conditional on the sequence of variables entering the model, i.e., the post-selection law of the knots of the LAR. Based on this result, we prove an exact of the False Discovery Rate (FDR) in the orthogonal design case and an exact control of the existence of false negatives in the general design case. First, we build a sequence of testing procedures on the variables entering the model and we give an exact control of the FDR in the orthogonal design case when the noise level can be unknown. Second, we introduce a new exact testing procedure on the existence of false negatives when the noise level can be unknown. This testing procedure can be deployed after any support selection procedure that will produce an estimation of the support (i.e., the indexes of nonzero coefficients) for any designs. The type~$I$ error of the test can be exactly controlled as long as the selection procedure follows some elementary hypotheses, referred to as admissible selection procedures. These support selection procedures are such that the estimation of the support is given by the $k$ first variables entering the model where the random variable $k$ is a stopping time. Monte-Carlo simulations and a real data experiment are provided to illustrate our results.

With the increase in the amount of data and the expansion of model scale, distributed parallel training becomes an important and successful technique to address the optimization challenges. Nevertheless, although distributed stochastic gradient descent (SGD) algorithms can achieve a linear iteration speedup, they are limited significantly in practice by the communication cost, making it difficult to achieve a linear time speedup. In this paper, we propose a computation and communication decoupled stochastic gradient descent (CoCoD-SGD) algorithm to run computation and communication in parallel to reduce the communication cost. We prove that CoCoD-SGD has a linear iteration speedup with respect to the total computation capability of the hardware resources. In addition, it has a lower communication complexity and better time speedup comparing with traditional distributed SGD algorithms. Experiments on deep neural network training demonstrate the significant improvements of CoCoD-SGD: when training ResNet18 and VGG16 with 16 Geforce GTX 1080Ti GPUs, CoCoD-SGD is up to 2-3$\times$ faster than traditional synchronous SGD.

In complex tasks, such as those with large combinatorial action spaces, random exploration may be too inefficient to achieve meaningful learning progress. In this work, we use a curriculum of progressively growing action spaces to accelerate learning. We assume the environment is out of our control, but that the agent may set an internal curriculum by initially restricting its action space. Our approach uses off-policy reinforcement learning to estimate optimal value functions for multiple action spaces simultaneously and efficiently transfers data, value estimates, and state representations from restricted action spaces to the full task. We show the efficacy of our approach in proof-of-concept control tasks and on challenging large-scale StarCraft micromanagement tasks with large, multi-agent action spaces.

Physical modeling of robotic system behavior is the foundation for controlling many robotic mechanisms to a satisfactory degree. Mechanisms are also typically designed in a way that good model accuracy can be achieved with relatively simple models and model identification strategies. If the modeling accuracy using physically based models is not enough or too complex, model-free methods based on machine learning techniques can help. Of particular interest to us was therefore the question to what degree semi-parametric modeling techniques, meaning combinations of physical models with machine learning, increase the modeling accuracy of inverse dynamics models which are typically used in robot control. To this end, we evaluated semi-parametric Gaussian process regression and a novel model-based neural network architecture, and compared their modeling accuracy to a series of naive semi-parametric, parametric-only and non-parametric-only regression methods. The comparison has been carried out on three test scenarios, one involving a real test-bed and two involving simulated scenarios, with the most complex scenario targeting the modeling a simulated robot's inverse dynamics model. We found that in all but one case, semi-parametric Gaussian process regression yields the most accurate models, also with little tuning required for the training procedure.

Membership inference (MI) attacks exploit a learned model's lack of generalization to infer whether a given sample was in the model's training set. Known MI attacks generally work by casting the attacker's goal as a supervised learning problem, training an attack model from predictions generated by the target model, or by others like it. However, we find that these attacks do not often provide a meaningful basis for confidently inferring training set membership, as the attack models are not well-calibrated. Moreover, these attacks do not significantly outperform a trivial attack that predicts that a point is a member if and only if the model correctly predicts its label. In this work we present well-calibrated MI attacks that allow the attacker to accurately control the minimum confidence with which positive membership inferences are made. Our attacks take advantage of white-box information about the target model and leverage new insights about how overfitting occurs in deep neural networks; namely, we show how a model's idiosyncratic use of features can provide evidence for membership. Experiments on seven real-world datasets show that our attacks support calibration for high-confidence inferences, while outperforming previous MI attacks in terms of accuracy. Finally, we show that our attacks achieve non-trivial advantage on some models with low generalization error, including those trained with small-epsilon-differential privacy; for large-epsilon (epsilon=16, as reported in some industrial settings), the attack performs comparably to unprotected models.