In many applications where collecting data is expensive, for example neuroscience or medical imaging, the sample size is typically small compared to the feature dimension. It is challenging in this setting to train expressive, non-linear models without overfitting. These datasets call for intelligent regularization that exploits known structure, such as correlations between the features arising from the measurement device. However, existing structured regularizers need specially crafted solvers, which are difficult to apply to complex models. We propose a new regularizer specifically designed to leverage structure in the data in a way that can be applied efficiently to complex models. Our approach relies on feature grouping, using a fast clustering algorithm inside a stochastic gradient descent loop: given a family of feature groupings that capture feature covariations, we randomly select these groups at each iteration. We show that this approach amounts to enforcing a denoising regularizer on the solution. The method is easy to implement in many model architectures, such as fully connected neural networks, and has a linear computational cost. We apply this regularizer to a real-world fMRI dataset and the Olivetti Faces datasets. Experiments on both datasets demonstrate that the proposed approach produces models that generalize better than those trained with conventional regularizers, and also improves convergence speed.

As algorithmic prediction systems have become widespread, fears that these systems may inadvertently discriminate against members of underrepresented populations have grown. With the goal of understanding fundamental principles that underpin the growing number of approaches to mitigating algorithmic discrimination, we investigate the role of information in fair prediction. A common strategy for decision-making uses a predictor to assign individuals a risk score; then, individuals are selected or rejected on the basis of this score. In this work, we formalize a framework for measuring the information content of predictors. Central to this framework is the notion of a refinement; intuitively, a refinement of a predictor $z$ increases the overall informativeness of the predictions without losing the information already contained in $z$. We show that increasing information content through refinements improves the downstream selection rules across a wide range of fairness measures (e.g. true positive rates, false positive rates, selection rates). In turn, refinements provide a simple but effective tool for reducing disparity in treatment and impact without sacrificing the utility of the predictions. Our results suggest that in many applications, the perceived "cost of fairness" results from an information disparity across populations, and thus, may be avoided with improved information.

We formulate a new class of conditional generative models based on probability flows. Trained with maximum likelihood, it provides efficient inference and sampling from class-conditionals or the joint distribution, and does not require a priori knowledge of the number of classes or the relationships between classes. This allows one to train generative models from multiple, heterogeneous datasets, while retaining strong prior models over subsets of the data (e.g., from a single dataset, class label, or attribute). In this paper, in addition to end-to-end learning, we show how one can learn a single model from multiple datasets with a relatively weak Glow architecture, and then extend it by conditioning on different knowledge types (e.g., a single dataset). This yields log likelihood comparable to state-of-the-art, compelling samples from conditional priors.

Machine-learning-based data-driven applications have become ubiquitous, e.g., health-care analysis and database system optimization. Big training data and large (deep) models are crucial for good performance. Dropout has been widely used as an efficient regularization technique to prevent large models from overfitting. However, many recent works show that dropout does not bring much performance improvement for deep convolutional neural networks (CNNs), a popular deep learning model for data-driven applications. In this paper, we formulate existing dropout methods for CNNs under the same analysis framework to investigate the failures. We attribute the failure to the conflicts between the dropout and the batch normalization operation after it. Consequently, we propose to change the order of the operations, which results in new building blocks of CNNs.Extensive experiments on benchmark datasets CIFAR, SVHN and ImageNet have been conducted to compare the existing building blocks and our new building blocks with different dropout methods. The results confirm the superiority of our proposed building blocks due to the regularization and implicit model ensemble effect of dropout. In particular, we improve over state-of-the-art CNNs with significantly better performance of 3.17%, 16.15%, 1.44%, 21.46% error rate on CIFAR-10, CIFAR-100, SVHN and ImageNet respectively.

Fall detection systems are concerned with rapidly detecting the occurrence of falls from elderly and disabled users using data from a body-worn inertial measurement unit (IMU), which is typically used in conjunction with machine learning-based classification. Such systems, however, necessitate the collection of high-resolution measurements that can violate users' privacy, such as revealing their gait, activities of daily living (ADLs), and relative position using dead reckoning. In this paper, for the first time, we present the design, implementation and evaluation of applying multi-party computation (MPC) to IMU-based fall detection for assuring the confidentiality of device measurements. The system is evaluated in a cloud-based setting that precludes parties from learning the underlying data using three parties deployed in geographically disparate locations in three cloud configurations. Using a publicly-available dataset comprising fall data from real-world users, we explore the applicability of derivative-based features to mitigate the complexity of MPC-based operations in a state-of-the-art fall detection system. We demonstrate that MPC-based fall detection from IMU measurements is both feasible and practical, executing in 365.2 milliseconds, which falls well below the required time window for on-device data acquisition (750ms).

Shapelet is a discriminative subsequence of time series. An advanced shapelet-based method is to embed shapelet into accurate and fast random forest. However, it shows several limitations. First, random shapelet forest requires a large training cost for split threshold searching. Second, a single shapelet provides limited information for only one branch of the decision tree, resulting in insufficient accuracy and interpretability. Third, randomized ensemble causes interpretability declining. For that, this paper presents Random Pairwise Shapelets Forest (RPSF). RPSF combines a pair of shapelets from different classes to construct random forest. It omits threshold searching to be more efficient, includes more information for each node of the forest to be more effective. Moreover, a discriminability metric, Decomposed Mean Decrease Impurity (DMDI), is proposed to identify influential region for every class. Extensive experiments show RPSF improves the accuracy and training speed of shapelet-based forest. Case studies demonstrate the interpretability of our method.

Recently, deep learning has become a de facto standard in machine learning with convolutional neural networks (CNNs) demonstrating spectacular success on a wide variety of tasks. However, CNNs are typically very demanding computationally at inference time. One of the ways to alleviate this burden on certain hardware platforms is quantization relying on the use of low-precision arithmetic representation for the weights and the activations. Another popular method is the pruning of the number of filters in each layer. While mainstream deep learning methods train the neural networks weights while keeping the network architecture fixed, the emerging neural architecture search (NAS) techniques make the latter also amenable to training. In this paper, we formulate optimal arithmetic bit length allocation and neural network pruning as a NAS problem, searching for the configurations satisfying a computational complexity budget while maximizing the accuracy. We use a differentiable search method based on the continuous relaxation of the search space proposed by Liu et al. (arXiv:1806.09055). We show, by grid search, that heterogeneous quantized networks suffer from a high variance which renders the benefit of the search questionable. For pruning, improvement over homogeneous cases is possible, but it is still challenging to find those configurations with the proposed method. The code is publicly available at https://github.com/yochaiz/Slimmable and https://github.com/yochaiz/darts-UNIQ .

Several methods of estimating the mutual information of random variables have been developed in recent years. They can prove valuable for novel approaches to learning statistically independent features. In this paper, we use one of these methods, a mutual information neural estimation (MINE) network, to present a proof-of-concept of how a neural network can perform linear ICA. We minimize the mutual information, as estimated by a MINE network, between the output units of a differentiable encoder network. This is done by simple alternate optimization of the two networks. The method is shown to get a qualitatively equal solution to FastICA on blind-source-separation of noisy sources.

In low-rank approximation with missing entries, given $A\in \mathbb{R}^{n\times n}$ and binary $W \in \{0,1\}^{n\times n}$, the goal is to find a rank-$k$ matrix $L$ for which: $$cost(L)=\sum_{i=1}^{n} \sum_{j=1}^{n}W_{i,j}\cdot (A_{i,j} - L_{i,j})^2\le OPT+\epsilon \|A\|_F^2,$$ where $OPT=\min_{rank-k\ \hat{L}}cost(\hat L)$. This problem is also known as matrix completion and, depending on the choice of $W$, captures low-rank plus diagonal decomposition, robust PCA, low-rank recovery from monotone missing data, and a number of other important problems. Many of these problems are NP-hard, and while algorithms with provable guarantees are known in some cases, they either 1) run in time $n^{\Omega(k^2/\epsilon)}$, or 2) make strong assumptions, e.g., that $A$ is incoherent or that $W$ is random. In this work, we consider $bicriteria\ algorithms$, which output $L$ with rank $k' > k$. We prove that a common heuristic, which simply sets $A$ to $0$ where $W$ is $0$, and then computes a standard low-rank approximation, achieves the above approximation bound with rank $k'$ depending on the $communication\ complexity$ of $W$. Namely, interpreting $W$ as the communication matrix of a Boolean function $f(x,y)$ with $x,y\in \{0,1\}^{\log n}$, it suffices to set $k'=O(k\cdot 2^{R^{1-sided}_{\epsilon}(f)})$, where $R^{1-sided}_{\epsilon}(f)$ is the randomized communication complexity of $f$ with $1$-sided error probability $\epsilon$. For many problems, this yields bicriteria algorithms with $k'=k\cdot poly((\log n)/\epsilon)$. We prove a similar bound using the randomized communication complexity with $2$-sided error. Further, we show that different models of communication yield algorithms for natural variants of the problem. E.g., multi-player communication complexity connects to tensor decomposition and non-deterministic communication complexity to Boolean low-rank factorization.

The concern of potential privacy violation has prevented efficient use of big data for improving deep learning based applications. In this paper, we propose Morphed Learning, a privacy-preserving technique for deep learning based on data morphing that, allows data owners to share their data without leaking sensitive privacy information. Morphed Learning allows the data owners to send securely morphed data and provides the server with an Augmented Convolutional layer to train the network on morphed data without performance loss. Morphed Learning has these three features: (1) Strong protection against reverse-engineering on the morphed data; (2) Acceptable computational and data transmission overhead with no correlation to the depth of the neural network; (3) No degradation of the neural network performance. Theoretical analyses on CIFAR-10 dataset and VGG-16 network show that our method is capable of providing 10^89 morphing possibilities with only 5% computational overhead and 10% transmission overhead under limited knowledge attack scenario. Further analyses also proved that our method can offer same resilience against full knowledge attack if more resources are provided.

We consider the problem where an agent wants to find a hidden object that is randomly located in some vertex of a directed acyclic graph (DAG) according to a fixed but possibly unknown distribution. The agent can only examine vertices whose in-neighbors have already been examined. In this paper, we address a learning setting where we allow the agent to stop before having found the object and restart searching on a new independent instance of the same problem. Our goal is to maximize the total number of hidden objects found given a time budget. The agent can thus skip an instance after realizing that it would spend too much time on it. Our contributions are both to the search theory and multi-armed bandits. If the distribution is known, we provide a quasi-optimal and efficient stationary strategy. If the distribution is unknown, we additionally show how to sequentially approximate it and, at the same time, act near-optimally in order to collect as many hidden objects as possible.

We present cyber-security problems of high importance. We show that in order to solve these cyber-security problems, one must cope with certain machine learning challenges. We provide novel data sets representing the problems in order to enable the academic community to investigate the problems and suggest methods to cope with the challenges. We also present a method to generate labels via pivoting, providing a solution to common problems of lack of labels in cyber-security.

In recent years, Generative Adversarial Networks (GANs) have drawn a lot of attentions for learning the underlying distribution of data in various applications. Despite their wide applicability, training GANs is notoriously difficult. This difficulty is due to the min-max nature of the resulting optimization problem and the lack of proper tools of solving general (non-convex, non-concave) min-max optimization problems. In this paper, we try to alleviate this problem by proposing a new generative network that relies on the use of random discriminators instead of adversarial design. This design helps us to avoid the min-max formulation and leads to an optimization problem that is stable and could be solved efficiently. The performance of the proposed method is evaluated using handwritten digits (MNIST) and Fashion products (Fashion-MNIST) data sets. While the resulting images are not as sharp as adversarial training, the use of random discriminator leads to a much faster algorithm as compared to the adversarial counterpart. This observation, at the minimum, illustrates the potential of the random discriminator approach for warm-start in training GANs.

It is conventional wisdom in machine learning and data mining that logical models such as rule sets are more interpretable than other models, and that among such rule-based models, simpler models are more interpretable than more complex ones. In this position paper, we question this latter assumption by focusing on one particular aspect of interpretability, namely the plausibility of models. Roughly speaking, we equate the plausibility of a model with the likeliness that a user accepts it as an explanation for a prediction. In particular, we argue that, all other things being equal, longer explanations may be more convincing than shorter ones, and that the predominant bias for shorter models, which is typically necessary for learning powerful discriminative models, may not be suitable when it comes to user acceptance of the learned models. To that end, we first recapitulate evidence for and against this postulate, and then report the results of an evaluation in a crowd-sourcing study based on about 3.000 judgments. The results do not reveal a strong preference for simple rules, whereas we can observe a weak preference for longer rules in some domains. We then relate these results to well-known cognitive biases such as the conjunction fallacy, the representative heuristic, or the recogition heuristic, and investigate their relation to rule length and plausibility.

While Bayesian methods are extremely popular in statistics and machine learning, their application to massive datasets is often challenging, when possible at all. Indeed, the classical MCMC algorithms are prohibitively slow when both the model dimension and the sample size are large. Variational Bayesian methods aim at approximating the posterior by a distribution in a tractable family. Thus, MCMC are replaced by an optimization algorithm which is orders of magnitude faster. VB methods have been applied in such computationally demanding applications as including collaborative filtering, image and video processing, NLP and text processing... However, despite very nice results in practice, the theoretical properties of these approximations are usually not known. In this paper, we propose a general approach to prove the concentration of variational approximations of fractional posteriors. We apply our theory to two examples: matrix completion, and Gaussian VB.

Top