arXiv subjects

Ran Gilad-Bachrach

Publications and source records attributed to Ran Gilad-Bachrach.

30 records · Page 2Linked to original sources

arXiv2020

Algorithmic Copywriting: Automated Generation of Health-Related Advertisements to Improve their Performance

Search advertising, a popular method for online marketing, has been employed to improve health by eliciting positive behavioral change. However, writing effective advertisements requires expertise and experimentation, which may not be available to health authorities wishing to elicit such changes, especially when dealing with public health crises such as epidemic outbreaks. Here we develop a framework, comprised of two neural networks models, that automatically generate ads. First, it employs a generator model, which create ads from web pages. It then employs a translation model, which transcribes ads to improve performance. We trained the networks using 114K health-related ads shown on Microsoft Advertising. We measure ads performance using the click-through rates (CTR). Our experiments show that the generated advertisements received approximately the same CTR as human-authored ads. The marginal contribution of the generator model was, on average, 28\% lower than that of human-authored ads, while the translator model received, on average, 32\% more clicks than human-authored ads. Our analysis shows that the translator model produces ads reflecting higher values of psychological attributes associated with a user action, including higher valance and arousal, and more calls-to-actions. In contrast, levels of these attributes in ads produced by the generator model are similar to those of human-authored ads. Our results demonstrate the ability to automatically generate useful advertisements for the health domain. We believe that our work offers health authorities an improved ability to nudge people towards healthier behaviors while saving the time and cost needed to build effective advertising campaigns.

cs.IR↗

arXiv2020

Smooth Sensitivity Based Approach for Differentially Private Principal Component Analysis

Currently known methods for this task either employ the computationally intensive \emph{exponential mechanism} or require an access to the covariance matrix, and therefore fail to utilize potential sparsity of the data. The problem of designing simpler and more efficient methods for this task has been raised as an open problem in \cite{kapralov2013differentially}. In this paper we address this problem by employing the output perturbation mechanism. Despite being arguably the simplest and most straightforward technique, it has been overlooked due to the large \emph{global sensitivity} associated with publishing the leading eigenvector. We tackle this issue by adopting a \emph{smooth sensitivity} based approach, which allows us to establish differential privacy (in a worst-case manner) and near-optimal sample complexity results under eigengap assumption. We consider both the pure and the approximate notions of differential privacy, and demonstrate a tradeoff between privacy level and sample complexity. We conclude by suggesting how our results can be extended to related problems.

cs.LG↗

arXiv2019

Low Latency Privacy Preserving Inference

When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity. Recent studies combined Homomorphic Encryption with neural networks to make inferences while protecting against information leakage. However, these methods are limited by the width and depth of neural networks that can be used (and hence the accuracy) and exhibit high latency even for relatively simple networks. In this study we provide two solutions that address these limitations. In the first solution, we present more than $10\times$ improvement in latency and enable inference on wider networks compared to prior attempts with the same level of security. The improved performance is achieved by novel methods to represent the data during the computation. In the second solution, we apply the method of transfer learning to provide private inference services using deep networks with latency of $\sim0.16$ seconds. We demonstrate the efficacy of our methods on several computer vision tasks.

cs.LG↗

arXiv2019

E-Gotsky: Sequencing Content using the Zone of Proximal Development

Vygotsky's notions of Zone of Proximal Development and Dynamic Assessment emphasize the importance of personalized learning that adapts to the needs and abilities of the learners and enables more efficient learning. In this work we introduce a novel adaptive learning engine called E-gostky that builds on these concepts to personalize the learning path within an e-learning system. E-gostky uses machine learning techniques to select the next content item that will challenge the student but will not be overwhelming, keeping students in their Zone of Proximal Development. To evaluate the system, we conducted an experiment where hundreds of students from several different elementary schools used our engine to learn fractions for five months. Our results show that using E-gostky can significantly reduce the time required to reach similar mastery. Specifically, in our experiment, it took students who were using the adaptive learning engine $17\%$ less time to reach a similar level of mastery as of those who didn't. Moreover, students made greater efforts to find the correct answer rather than guessing and class teachers reported that even students with learning disabilities showed higher engagement.

cs.HC↗

arXiv2018

Turning Lemons into Peaches using Secure Computation

In many cases, assessing the quality of goods is hard. For example, when purchasing a car, it is hard to measure how pollutant the car is since there are infinitely many driving conditions to be tested. Typically, these situations are considered under the umbrella of information asymmetry and as Akelrof showed may lead to a market of lemons. However, we argue that in many of these situations, the problem is not the missing information but the computational challenge of obtaining it. In a nut-shell, if verifying the value of goods requires a large amount of computation or even infinite amounts of computation, the buyer is forced to use a finite test that samples, in some sense, the quality of the goods. However, if the seller knows the test, then the seller can over-fit the test and create goods that pass the quality test despite not having the desired quality. We show different solutions to this situation including a novel approach that uses secure computation to hide the test from the seller to prevent over-fitting.

cs.GT↗

arXiv2018

Modeling and Simultaneously Removing Bias via Adversarial Neural Networks

In real world systems, the predictions of deployed Machine Learned models affect the training data available to build subsequent models. This introduces a bias in the training data that needs to be addressed. Existing solutions to this problem attempt to resolve the problem by either casting this in the reinforcement learning framework or by quantifying the bias and re-weighting the loss functions. In this work, we develop a novel Adversarial Neural Network (ANN) model, an alternative approach which creates a representation of the data that is invariant to the bias. We take the Paid Search auction as our working example and ad display position features as the confounding features for this setting. We show the success of this approach empirically on both synthetic data as well as real world paid search auction data from a major search engine.

cs.LG↗

arXiv2015

DART: Dropouts meet Multiple Additive Regression Trees

Multiple Additive Regression Trees (MART), an ensemble model of boosted regression trees, is known to deliver high prediction accuracy for diverse tasks, and it is widely used in practice. However, it suffers an issue which we call over-specialization, wherein trees added at later iterations tend to impact the prediction of only a few instances, and make negligible contribution towards the remaining instances. This negatively affects the performance of the model on unseen data, and also makes the model over-sensitive to the contributions of the few, initially added tress. We show that the commonly used tool to address this issue, that of shrinkage, alleviates the problem only to a certain extent and the fundamental issue of over-specialization still remains. In this work, we explore a different approach to address the problem that of employing dropouts, a tool that has been recently proposed in the context of learning deep neural networks. We propose a novel way of employing dropouts in MART, resulting in the DART algorithm. We evaluate DART on ranking, regression and classification tasks, using large scale, publicly available datasets, and show that DART outperforms MART in each of the tasks, with a significant margin. We also show that DART overcomes the issue of over-specialization to a considerable extent.

cs.LG↗

arXiv2014

Crypto-Nets: Neural Networks over Encrypted Data

The problem we address is the following: how can a user employ a predictive model that is held by a third party, without compromising private information. For example, a hospital may wish to use a cloud service to predict the readmission risk of a patient. However, due to regulations, the patient's medical files cannot be revealed. The goal is to make an inference using the model, without jeopardizing the accuracy of the prediction or the privacy of the data. To achieve high accuracy, we use neural networks, which have been shown to outperform other learning models for many tasks. To achieve the privacy requirements, we use homomorphic encryption in the following protocol: the data owner encrypts the data and sends the ciphertexts to the third party to obtain a prediction from a trained model. The model operates on these ciphertexts and sends back the encrypted prediction. In this protocol, not only the data remains private, even the values predicted are available only to the data owner. Using homomorphic encryption and modifications to the activation functions and training algorithms of neural networks, we show that it is protocol is possible and may be feasible. This method paves the way to build a secure cloud-based neural network prediction services without invading users' privacy.

cs.LG↗

arXiv2013

Using Multiple Samples to Learn Mixture Models

In the mixture models problem it is assumed that there are $K$ distributions $θ_{1},\ldots,θ_{K}$ and one gets to observe a sample from a mixture of these distributions with unknown coefficients. The goal is to associate instances with their generating distributions, or to identify the parameters of the hidden distributions. In this work we make the assumption that we have access to several samples drawn from the same $K$ underlying distributions, but with different mixing weights. As with topic modeling, having multiple samples is often a reasonable assumption. Instead of pooling the data into one sample, we prove that it is possible to use the differences between the samples to better recover the underlying structure. We present algorithms that recover the underlying structure under milder assumptions than the current state of art when either the dimensionality or the separation is high. The methods, when applied to topic modeling, allow generalization to words not present in the training data.

stat.ML↗

arXiv2012

Optimal Distributed Online Prediction using Mini-Batches

Online prediction methods are typically presented as serial algorithms running on a single processor. However, in the age of web-scale prediction problems, it is increasingly common to encounter situations where a single processor cannot keep up with the high rate at which inputs arrive. In this work, we present the \emph{distributed mini-batch} algorithm, a method of converting many serial gradient-based online prediction algorithms into distributed algorithms. We prove a regret bound for this method that is asymptotically optimal for smooth convex loss functions and stochastic inputs. Moreover, our analysis explicitly takes into account communication latencies between nodes in the distributed environment. We show how our method can be used to solve the closely-related distributed stochastic optimization problem, achieving an asymptotically linear speed-up over multiple processors. Finally, we demonstrate the merits of our approach on a web-scale online prediction problem.

cs.LG↗

arXiv2010

Robust Distributed Online Prediction

The standard model of online prediction deals with serial processing of inputs by a single processor. However, in large-scale online prediction problems, where inputs arrive at a high rate, an increasingly common necessity is to distribute the computation across several processors. A non-trivial challenge is to design distributed algorithms for online prediction, which maintain good regret guarantees. In \cite{DMB}, we presented the DMB algorithm, which is a generic framework to convert any serial gradient-based online prediction algorithm into a distributed algorithm. Moreover, its regret guarantee is asymptotically optimal for smooth convex loss functions and stochastic inputs. On the flip side, it is fragile to many types of failures that are common in distributed environments. In this companion paper, we present variants of the DMB algorithm, which are resilient to many types of network failures, and tolerant to varying performance of the computing nodes.

cs.LG↗

arXiv2009

Efficient Human Computation

Collecting large labeled data sets is a laborious and expensive task, whose scaling up requires division of the labeling workload between many teachers. When the number of classes is large, miscorrespondences between the labels given by the different teachers are likely to occur, which, in the extreme case, may reach total inconsistency. In this paper we describe how globally consistent labels can be obtained, despite the absence of teacher coordination, and discuss the possible efficiency of this process in terms of human labor. We define a notion of label efficiency, measuring the ratio between the number of globally consistent labels obtained and the number of labels provided by distributed teachers. We show that the efficiency depends critically on the ratio alpha between the number of data instances seen by a single teacher, and the number of classes. We suggest several algorithms for the distributed labeling problem, and analyze their efficiency as a function of alpha. In addition, we provide an upper bound on label efficiency for the case of completely uncoordinated teachers, and show that efficiency approaches 0 as the ratio between the number of labels each teacher provides and the number of classes drops (i.e. alpha goes to 0).

cs.LG↗