Search

Non-proportional hazards data are routinely encountered in randomized clinical trials. In such cases, classic Cox proportional hazards model can suffer from severe power loss, with difficulty in interpretation of the estimated hazard ratio since the treatment effect varies over time. We propose CauchyCP, an omnibus test of change-point Cox regression models, to overcome both challenges while detecting signals of non-proportional hazards patterns. Extensive simulation studies demonstrate that, compared to existing treatment comparison tests under non-proportional hazards, the proposed CauchyCP test 1) controls the type I error better at small $\alpha$ levels ($< 0.01$); 2) increases the power of detecting time-varying effects; and 3) is more computationally efficient. The superior performance of CauchyCP is further illustrated using retrospective analyses of two randomized clinical trial datasets and a pharmacogenetic biomarker study dataset. The R package $\textit{CauchyCP}$ is publicly available on CRAN.

Code clones are duplicate code fragments that share (nearly) similar syntax or semantics. Code clone detection plays an important role in software maintenance, code refactoring, and reuse. A substantial amount of research has been conducted in the past to detect clones. A majority of these approaches use lexical and syntactic information to detect clones. However, only a few of them target semantic clones. Recently, motivated by the success of deep learning models in other fields, including natural language processing and computer vision, researchers have attempted to adopt deep learning techniques to detect code clones. These approaches use lexical information (tokens) and(or) syntactic structures like abstract syntax trees (ASTs) to detect code clones. However, they do not make sufficient use of the available structural and semantic information hence, limiting their capabilities. This paper addresses the problem of semantic code clone detection using program dependency graphs and geometric neural networks, leveraging the structured syntactic and semantic information. We have developed a prototype tool HOLMES, based on our novel approach, and empirically evaluated it on popular code clone benchmarks. Our results show that HOLMES performs considerably better than the other state-of-the-art tool, TBCCD. We also evaluated HOLMES on unseen projects and performed cross dataset experiments to assess the generalizability of HOLMES. Our results affirm that HOLMES outperforms TBCCD since most of the pairs that HOLMES detected were either undetected or suboptimally reported by TBCCD., Comment: Under Review IEEE Transactions on Software Engineering

Subset selection algorithms are ubiquitous in AI-driven applications, including, online recruiting portals and image search engines, so it is imperative that these tools are not discriminatory on the basis of protected attributes such as gender or race. Currently, fair subset selection algorithms assume that the protected attributes are known as part of the dataset. However, protected attributes may be noisy due to errors during data collection or if they are imputed (as is often the case in real-world settings). While a wide body of work addresses the effect of noise on the performance of machine learning algorithms, its effect on fairness remains largely unexamined. We find that in the presence of noisy protected attributes, in attempting to increase fairness without considering noise, one can, in fact, decrease the fairness of the result! Towards addressing this, we consider an existing noise model in which there is probabilistic information about the protected attributes (e.g., [58, 34, 20, 46]), and ask is fair selection possible under noisy conditions? We formulate a ``denoised'' selection problem which functions for a large class of fairness metrics; given the desired fairness goal, the solution to the denoised problem violates the goal by at most a small multiplicative amount with high probability. Although this denoised problem turns out to be NP-hard, we give a linear-programming based approximation algorithm for it. We evaluate this approach on both synthetic and real-world datasets. Our empirical results show that this approach can produce subsets which significantly improve the fairness metrics despite the presence of noisy protected attributes, and, compared to prior noise-oblivious approaches, has better Pareto-tradeoffs between utility and fairness.

The emergence of Human Body Communication (HBC) as an alternative to wireless body area networks (WBAN) has led to the development of small sized, energy efficient and more secure wearable and implantable devices forming a network in and around the body. Previous studies claim that though HBC is comparatively more secure than WBAN, nevertheless, the electromagnetic (EM) radiative nature of HBC in >10MHz region makes the information susceptible to eavesdropping. Furthermore, interferences may be picked up by the body due to the human body antenna effect in the 40-400MHz range. Alternatively, electro-quasistatic (EQS) mode of HBC forms an attractive way for covert data transmission in the sub 10MHz region by allowing the signal to be contained within the body. However, there is a gap in the knowledge about the mechanism and sources of interference in this region (crucial in allowing for proper choice of data transmission band). In this paper, the interference coupling modality in the EQS region is explained along with its possible sources. Interferences seen by the wearable in the actual scenario is a non-trivial problem and a suitable measurement EQS HBC setup is designed to recreate it by employing a wearable sized measurement setup having a small ground plane. For the first time, a human biophysical interference pickup model is proposed and interference measurement results using a wearable device are presented up to 250kHz in different environmental settings., Comment: This work has been submitted to the IEEE for possible publication

A robust body of evidence demonstrates the adverse effects of implicit bias in various contexts--from hiring to health care. The Rooney Rule is an intervention developed to counter implicit bias and has been implemented in the private and public sectors. The Rooney Rule requires that a selection panel include at least one candidate from an underrepresented group in their shortlist of candidates. Recently, Kleinberg and Raghavan proposed a model of implicit bias and studied the effectiveness of the Rooney Rule when applied to a single selection decision. However, selection decisions often occur repeatedly over time. Further, it has been observed that, given consistent counterstereotypical feedback, implicit biases against underrepresented candidates can change. We consider a model of how a selection panel's implicit bias changes over time given their hiring decisions either with or without the Rooney Rule in place. Our main result is that, when the panel is constrained by the Rooney Rule, their implicit bias roughly reduces at a rate that is the inverse of the size of the shortlist--independent of the number of candidates, whereas without the Rooney Rule, the rate is inversely proportional to the number of candidates. Thus, when the number of candidates is much larger than the size of the shortlist, the Rooney Rule enables a faster reduction in implicit bias, providing an additional reason in favor of using it as a strategy to mitigate implicit bias. Towards empirically evaluating the long-term effect of the Rooney Rule in repeated selection decisions, we conduct an iterative candidate selection experiment on Amazon MTurk. We observe that, indeed, decision-makers subject to the Rooney Rule select more minority candidates in addition to those required by the rule itself than they would if no rule is in effect, and do so without considerably decreasing the utility of candidates selected., Comment: Abstract shortened to satisfy the 1920 character limit

Static program analysis tools are often required to work with only a small part of a program's source code, either due to the unavailability of the entire program or the lack of need to analyze the complete code. This makes it challenging to use static analysis tools that require a complete and typed intermediate representation (IR). We present JCoffee, a tool that leverages compiler feedback to convert partial Java programs into their compilable counterparts by simulating the presence of missing surrounding code. It works with any well-typed code snippet (class, function, or even an unenclosed group of statements) while making minimal changes to the input code fragment. A demo of the tool is available here: https://youtu.be/O4h2g_n2Qls, Comment: 4 pages, 2 figures

Attorneys and others have a strong interest in having a digital library with suitable services (e.g., summarizing, searching, and browsing) to help them work with large corpora of legal depositions. Their needs often involve understanding the semantics of such documents. That depends in part on the role of the deponent, e.g., plaintiff, defendant, law enforcement personnel, expert, etc. In the case of tort litigation associated with property and casualty insurance claims, such as relating to an injury, it is important to know not only about liability, but also about events, accidents, physical conditions, and treatments. We hypothesize that a legal deposition consists of various aspects that are discussed as part of the deponent testimony. Accordingly, we developed an ontology of aspects in a legal deposition for accident and injury cases. Using that, we have developed a classifier that can identify portions of text for each of the aspects of interest. Doing so was complicated by the peculiarities of this genre, e.g., that deposition transcripts generally consist of data in the form of question-answer (QA) pairs. Accordingly, our automated system starts with pre-processing, and then transforms the QA pairs into a canonical form made up of declarative sentences. Classifying the declarative sentences that are generated, according to the aspect, can then help with downstream tasks such as summarization, segmentation, question-answering, and information retrieval. Our methods have achieved a classification F1 score of 0.83. Having the aspects classified with a good accuracy will help in choosing QA pairs that can be used as candidate summary sentences, and to generate an informative summary for legal professionals or insurance claim agents. Our methodology could be extended to legal depositions of other kinds, and to aid services like searching., Comment: 19 pages, 3 figures, 11 tables, detailed version of shorter paper being submitted to a conference

A randomized trial allows estimation of the causal effect of an intervention compared to a control in the overall population and in subpopulations defined by baseline characteristics. Often, however, clinical questions also arise regarding the treatment effect in subpopulations of patients, which would experience clinical or disease related events post-randomization. Events that occur after treatment initiation and potentially affect the interpretation or the existence of the measurements are called {\it intercurrent events} in the ICH E9(R1) guideline. If the intercurrent event is a consequence of treatment, randomization alone is no longer sufficient to meaningfully estimate the treatment effect. Analyses comparing the subgroups of patients without the intercurrent events for intervention and control will not estimate a causal effect. This is well known, but post-hoc analyses of this kind are commonly performed in drug development. An alternative approach is the principal stratum strategy, which classifies subjects according to their potential occurrence of an intercurrent event on both study arms. We illustrate with examples that questions formulated through principal strata occur naturally in drug development and argue that approaching these questions with the ICH E9(R1) estimand framework has the potential to lead to more transparent assumptions as well as more adequate analyses and conclusions. In addition, we provide an overview of assumptions required for estimation of effects in principal strata. Most of these assumptions are unverifiable and should hence be based on solid scientific understanding. Sensitivity analyses are needed to assess robustness of conclusions.

LPCNet is an efficient vocoder that combines linear prediction and deep neural network modules to keep the computational complexity low. In this work, we present two techniques to further reduce it's complexity, aiming for a low-cost LPCNet vocoder-based neural Text-to-Speech (TTS) System. These techniques are: 1) Sample-bunching, which allows LPCNet to generate more than one audio sample per inference; and 2) Bit-bunching, which reduces the computations in the final layer of LPCNet. With the proposed bunching techniques, LPCNet, in conjunction with a Deep Convolutional TTS (DCTTS) acoustic model, shows a 2.19x improvement over the baseline run-time when running on a mobile device, with a less than 0.1 decrease in TTS mean opinion score (MOS)., Comment: Interspeech 2020

Increasing demand for on-device Automatic Speech Recognition (ASR) systems has resulted in renewed interests in developing automatic model compression techniques. Past research have shown that AutoML-based Low Rank Factorization (LRF) technique, when applied to an end-to-end Encoder-Attention-Decoder style ASR model, can achieve a speedup of up to 3.7x, outperforming laborious manual rank-selection approaches. However, we show that current AutoML-based search techniques only work up to a certain compression level, beyond which they fail to produce compressed models with acceptable word error rates (WER). In this work, we propose an iterative AutoML-based LRF approach that achieves over 5x compression without degrading the WER, thereby advancing the state-of-the-art in ASR compression.

Users of music streaming, video streaming, news recommendation, and e-commerce services often engage with content in a sequential manner. Providing and evaluating good sequences of recommendations is therefore a central problem for these services. Prior reweighting-based counterfactual evaluation methods either suffer from high variance or make strong independence assumptions about rewards. We propose a new counterfactual estimator that allows for sequential interactions in the rewards with lower variance in an asymptotically unbiased manner. Our method uses graphical assumptions about the causal relationships of the slate to reweight the rewards in the logging policy in a way that approximates the expected sum of rewards under the target policy. Extensive experiments in simulation and on a live recommender system show that our approach outperforms existing methods in terms of bias and data efficiency for the sequential track recommendations problem.

The doped silicon regions (tubs) in PWFDSOI MOSFET cause significant reduction in OFF current by reducing the number of carriers contributing to the OFF current. The emphasis of the simulation and device physics study on PWFDSOI MOSFET presented in this paper is on the scalability of the device to 10 nm gate length and its related information. A high ION /IOFF ratio of 7.6 x 10^5 and subthreshold swing of 87 mV/decade were achieved in 10 nm gate length PWFDSOI MOSFET. The study was performed on devices with unstrained silicon channel.

A novel planar device having doped silicon regions (tubs) under the source and drain of an FDSOI MOSFET is reported at 20 nm gate length. The doped silicon regions result in formation of potential wells (PW) in the source and drain regions of FDSOI MOSFET and thus, the device being called as Potential Well Based FDSOI MOSFET (PWFDSOI MOSFET). Simulation and device physics study on PWFDSOI MOSFET showed reduction in the OFF current of the device by orders of magnitude. A low IOF F of 22 pA/um, high ION /IOF F ratio of 1.5 x 107 and subthreshold swing of 76 mV/decade were achieved in 20 nm gate length PWFDSOI MOSFET. The study was performed on devices with unstrained silicon channel.

There is strong interest in estimating how the magnitude of treatment effects of an intervention vary across sub-groups of the population of interest. In our paper, we propose a two-study approach to first propose and then test heterogeneous treatment effects. In Study 1, we use a large observational dataset to learn sub-groups with the most distinctive treatment-outcome relationships ('high/low-impact sub-groups'). We adopt a model-based recursive partitioning approach to propose the high/low impact sub-groups, and validate them by using sample-splitting. While the first study rules out noise, there is potential bias in our estimated heterogeneous treatment effects. Study 2 uses an experimental design, and here we classify our sample units based on sub-groups learned in Study 1. We then estimate treatment effects within each of the groups, thereby testing the causal hypotheses proposed in Study 1. Using patient claims data from the NBER MarketScan database, we apply our approach to estimate heterogeneous effects of a switch to a high-deductible health insurance plan on use of outpatient care by patients with a common chronic condition. We extend the method to non-parametrically learn the sub-groups in Study 1. We also compare the methods' performance to other state-of-the-art methods in the literature that make use only of the Study 2 data.

Despite extensive research on cryptography, secure and efficient query processing over outsourced data remains an open challenge. This paper continues along with the emerging trend in secure data processing that recognizes that the entire dataset may not be sensitive, and hence, non-sensitivity of data can be exploited to overcome limitations of existing encryption-based approaches. We, first, provide a new security definition, entitled partitioned data security for guaranteeing that the joint processing of non-sensitive data (in cleartext) and sensitive data (in encrypted form) does not lead to any leakage. Then, this paper proposes a new secure approach, entitled query binning (QB) that allows secure execution of queries over non-sensitive and sensitive parts of the data. QB maps a query to a set of queries over the sensitive and non-sensitive data in a way that no leakage will occur due to the joint processing over sensitive and non-sensitive data. In particular, we propose secure algorithms for selection, range, and join queries to be executed over encrypted sensitive and cleartext non-sensitive datasets. Interestingly, in addition to improving performance, we show that QB actually strengthens the security of the underlying cryptographic technique by preventing size, frequency-count, and workload-skew attacks., Comment: This version has been accepted in ACM Transactions on Management Information Systems. The final published version of this paper may differ from this accepted version. A preliminary version of this paper [arXiv:1812.09233] was accepted and presented in IEEE ICDE 2019

Contact tracing has emerged as one of the main mitigation strategies to prevent the spread of pandemics such as COVID-19. Recently, several efforts have been initiated to track individuals, their movements, and interactions using technologies, e.g., Bluetooth beacons, cellular data records, and smartphone applications. Such solutions are often intrusive, potentially violating individual privacy rights and are often subject to regulations (e.g., GDPR and CCPR) that mandate the need for opt-in policies to gather and use personal information. In this paper, we introduce Quest, a system that empowers organizations to observe individuals and spaces to implement policies for social distancing and contact tracing using WiFi connectivity data in a passive and privacy-preserving manner. The goal is to ensure the safety of employees and occupants at an organization, while protecting the privacy of all parties. Quest incorporates computationally- and information-theoretically-secure protocols that prevent adversaries from gaining knowledge of an individual's location history (based on WiFi data); it includes support for accurately identifying users who were in the vicinity of a confirmed patient, and then informing them via opt-in mechanisms. Quest supports a range of privacy-enabled applications to ensure adherence to social distancing, monitor the flow of people through spaces, identify potentially impacted regions, and raise exposure alerts. We describe the architecture, design choices, and implementation of the proposed security/privacy techniques in Quest. We, also, validate the practicality of Quest and evaluate it thoroughly via an actual campus-scale deployment at UC Irvine over a very large dataset of over 50M tuples.

Randomized clinical trials are often designed to assess whether a test treatment prolongs survival relative to a control treatment. Increased patient heterogeneity, while desirable for generalizability of results, can weaken the ability of common statistical approaches to detect treatment differences, potentially hampering the regulatory approval of safe and efficacious therapies. A novel solution to this problem is proposed. A list of baseline covariates that have the potential to be prognostic for survival under either treatment is pre-specified in the analysis plan. At the analysis stage, using all observed survival times but blinded to patient-level treatment assignment, 'noise' covariates are removed with elastic net Cox regression. The shortened covariate list is used by a conditional inference tree algorithm to segment the heterogeneous trial population into subpopulations of prognostically homogeneous patients (risk strata). After patient-level treatment unblinding, a treatment comparison is done within each formed risk stratum and stratum-level results are combined for overall statistical inference. The impressive power-boosting performance of our proposed 5-step stratified testing and amalgamation routine (5-STAR), relative to that of the logrank test and other common approaches that do not leverage inherently structured patient heterogeneity, is illustrated using a hypothetical and two real datasets along with simulation results. Furthermore, the importance of reporting stratum-level comparative treatment effects (time ratios from accelerated failure time model fits in conjunction with model averaging and, as needed, hazard ratios from Cox proportional hazard model fits) is highlighted as a potential enabler of personalized medicine. A fiveSTAR R package is available at https://github.com/rmarceauwest/fiveSTAR., Comment: 40 pages; 9 figures, 2 tables

Despite exciting progress on cryptography, secure and efficient query processing over outsourced data remains an open challenge. We develop a communication-efficient and information-theoretically secure system, entitled Obscure for aggregation queries with conjunctive or disjunctive predicates, using secret-sharing. Obscure is strongly secure (i.e., secure regardless of the computational-capabilities of an adversary) and prevents the network, as well as, the (adversarial) servers to learn the user's queries, results, or the database. In addition, Obscure provides additional security features, such as hiding access-patterns (i.e., hiding the identity of the tuple satisfying a query) and hiding query-patterns (i.e., hiding which two queries are identical). Also, Obscure does not require any communication between any two servers that store the secret-shared data before/during/after the query execution. Moreover, our techniques deal with the secret-shared data that is outsourced by a single or multiple database owners, as well as, allows a user, which may not be the database owner, to execute the query over secret-shared data. We further develop (non-mandatory) privacy-preserving result verification algorithms that detect malicious behaviors, and experimentally validate the efficiency of Obscure on large datasets, the size of which prior approaches of secret-sharing or multi-party computation systems have not scaled to., Comment: A preliminary version of this work was accepted in VLDB 2019. This version has been accepted in IEEE Transactions on Knowledge and Data Engineering (TKDE). The final published version of this paper may differ from this accepted version

This paper explores the data cleaning challenges that arise in using WiFi connectivity data to locate users to semantic indoor locations such as buildings, regions, rooms. WiFi connectivity data consists of sporadic connections between devices and nearby WiFi access points (APs), each of which may cover a relatively large area within a building. Our system, entitled semantic LOCATion cleanER (LOCATER), postulates semantic localization as a series of data cleaning tasks - first, it treats the problem of determining the AP to which a device is connected between any two of its connection events as a missing value detection and repair problem. It then associates the device with the semantic subregion (e.g., a conference room in the region) by postulating it as a location disambiguation problem. LOCATER uses a bootstrapping semi-supervised learning method for coarse localization and a probabilistic method to achieve finer localization. The paper shows that LOCATER can achieve significantly high accuracy at both the coarse and fine levels.

The dielectric properties of Butan-1-ol with acetic acid mixture have been studied in the frequency range 10 MHz to 20 GHz using time domain spectroscopy in the reflection mode, at various temperatures (i.e. on 288 K, 298 K, 308 K and 318 K) and at eleven different concentrations. Dielectric parameters such as static dielectric constant and relaxation time have been determined. Excess dielectric constant, Excess inverse relaxation time, Kirkwood correlation factor and the thermodynamic parameters ({\Delta}H & {\Delta}S) have been estimated using these dielectric parameters. A drastic change in molecular structure of Butan-1-ol with acetic acid mixture observed due to strong intermolecular association among the molecules of complex sysytem. Thermodynamic parameters explored to insight molecular structures changes of mixtures at different temperatures., Comment: pages:14,08 Figures,Tables 05

Current approaches of enforcing FGAC in Database Management Systems (DBMS) do not scale in scenarios when the number of policies are in the order of thousands. This paper identifies such a use case in the context of emerging smart spaces wherein systems may be required by legislation, such as Europe's GDPR and California's CCPA, to empower users to specify who may have access to their data and for what purposes. We present Sieve, a layered approach of implementing FGAC in existing database systems, that exploits a variety of it's features such as UDFs, index usage hints, query explain; to scale to large number of policies. Given a query, Sieve exploits it's context to filter the policies that need to be checked. Sieve also generates guarded expressions that saves on evaluation cost by grouping the policies and cuts the read cost by exploiting database indices. Our experimental results, on two DBMS and two different datasets, show that Sieve scales to large data sets and to large policy corpus thus supporting real-time access in applications including emerging smart environments., Comment: Extended version of the paper submitted to Very Large Data Bases (VLDB) and is now under review

This paper focuses on the new privacy challenges that arise in smart homes. Specifically, the paper focuses on inferring the user's activities -- which may, in turn, lead to the user's privacy -- via inferences through device activities and network traffic analysis. We develop techniques that are based on a cryptographically secure token circulation in a ring network consisting of smart home devices to prevent inferences from device activities, via device workflow, i.e., inferences from a coordinated sequence of devices' actuation. The solution hides the device activity and corresponding channel activities, and thus, preserve the individual's activities. We also extend our solution to deal with a large number of devices and devices that produce large-sized data by implementing parallel rings. Our experiments also evaluate the performance in terms of communication overheads of the proposed approach and the obtained privacy., Comment: This version has been accepted in ACM Transactions on Cyber-Physical Systems (TCPS). A preliminary version of this paper was accepted in ACM Conference on Data and Application Security and Privacy (CODASPY) 2019. arXiv admin note: substantial text overlap with arXiv:1901.08618

This paper presents a stochastic optimization model for allocating and sharing a critical resource in the case of a pandemic. The demand for different entities peaks at different times, and an initial inventory from a central agency is to be allocated. The entities (states) may share the critical resource with a different state under a risk-averse condition. The model is applied to study the allocation of ventilator inventory in the COVID-19 pandemic by the Federal Emergency Management Agency of the US Department of Homeland Security (FEMA) to different states in the US. Findings suggest that if less than 60% of the ventilator inventory is available for non-COVID-19 patients, FEMA's stockpile of 20,000 ventilators (as of 03/23/2020) would be nearly adequate to meet the projected needs. However, when more than 75% of the available ventilator inventory must be reserved for non-COVID-19 patients, various degrees of shortfall are expected. In an extreme case, where the demand is assumed to be concentrated in the top-most quartile of the forecast confidence interval, the total shortfall over the planning horizon (till 05/31/20) is about 28,500 ventilator days, with a peak shortfall of 2,700 ventilators on 04/12/20. The results also suggest that in the worse-than-average to severe demand scenario cases, NY requires between 7,600-9,200 additional ventilators for COVID-19 patients during its peak demand. However, between 400 to 2,000 of these ventilators can be given to a different state after the peak demand in NY has subsided.

The growing deployment of Internet of Things (IoT) systems aims to ease the daily life of end-users by providing several value-added services. However, IoT systems may capture and store sensitive, personal data about individuals in the cloud, thereby jeopardizing user-privacy. Emerging legislation, such as California's CalOPPA and GDPR in Europe, support strong privacy laws to protect an individual's data in the cloud. One such law relates to strict enforcement of data retention policies. This paper proposes a framework, entitled IoT Expunge that allows sensor data providers to store the data in cloud platforms that will ensure enforcement of retention policies. Additionally, the cloud provider produces verifiable proofs of its adherence to the retention policies. Experimental results on a real-world smart building testbed show that IoT Expunge imposes minimal overheads to the user to verify the data against data retention policies., Comment: This paper has been accepted in 10th ACM Conference on Data and Application Security and Privacy (CODASPY), 2020

The authors have withdrawn this paper due to an error in the proof of Lemma 3.4. -------------------------------------------------------------------------------------------- The authors have withdrawn this paper due to an error in the proof of Lemma 3.4z(Assadi, Chen, and Khanna, 2019) define a 4-player hidden-pointer-chasing ($\mathsf{HPC}^4$), and using it, give strong multi-pass lower bounds for graph problems in the streaming model of computation and a lower bound on the query complexity of sub-modular minimization. We present a two-player version ($\mathsf{HPC}^2$) of $\mathsf{HPC}^4$ that has matching communication complexity to $\mathsf{HPC}^4$. Our formulation allows us to lower bound its communication complexity with a simple direct-sum argument. Using this lower bound on the communication complexity of $\mathsf{HPC}^2$, we retain the streaming and query complexity lower bounds by (Assadi, Chen, and Khanna, 2019). Further, by giving reductions from $\mathsf{HPC}^2$, we prove new multi-pass space lower bounds for graph problems in turnstile streams. In particular, we show that any algorithm which computes the exact weight of the maximum weighted matching in an $n$-vertex graph requires $\tilde{O}(n^{2})$ space unless it makes $\omega(\log n)$ passes over the turnstile stream, and that any algorithm which computes the minimum $s\text{-}t$ distance in an $n$-vertex graph requires $n^{2-o(1)}$ space unless it makes $n^{\Omega(1)}$ passes over the turnstile stream. Our reductions can be modified to use $\mathsf{HPC}^4$ as well., Comment: The authors have withdrawn this paper due to an error in the proof of Lemma 3.4

Implicit bias is the unconscious attribution of particular qualities (or lack thereof) to a member from a particular social group (e.g., defined by gender or race). Studies on implicit bias have shown that these unconscious stereotypes can have adverse outcomes in various social contexts, such as job screening, teaching, or policing. Recently, (Kleinberg and Raghavan, 2018) considered a mathematical model for implicit bias and showed the effectiveness of the Rooney Rule as a constraint to improve the utility of the outcome for certain cases of the subset selection problem. Here we study the problem of designing interventions for the generalization of subset selection -- ranking -- that requires to output an ordered set and is a central primitive in various social and computational contexts. We present a family of simple and interpretable constraints and show that they can optimally mitigate implicit bias for a generalization of the model studied in (Kleinberg and Raghavan, 2018). Subsequently, we prove that under natural distributional assumptions on the utilities of items, simple, Rooney Rule-like, constraints can also surprisingly recover almost all the utility lost due to implicit biases. Finally, we augment our theoretical results with empirical findings on real-world distributions from the IIT-JEE (2009) dataset and the Semantic Scholar Research corpus., Comment: This paper will appear at the ACM FAT* 2020 conference

Understanding and learning the characteristics of network paths has been of particular interest for decades and has led to several successful applications. Such analysis becomes challenging for urban networks as their size and complexity are significantly higher compared to other networks. The state-of-the-art machine learning (ML) techniques allow us to detect hidden patterns and, thus, infer the features associated with them. However, very little is known about the impact on the performance of such predictive models by the use of different input representations. In this paper, we design and evaluate six different graph input representations (i.e., representations of the network paths), by considering the network's topological and temporal characteristics, for being used as inputs for machine learning models to learn the behavior of urban networks paths. The representations are validated and then tested with a real-world taxi journeys dataset predicting the tips using a road network of New York. Our results demonstrate that the input representations that use temporal information help the model to achieve the highest accuracy (RMSE of 1.42$).

Transportability provides a principled framework to address the problem of applying study results to new populations. Here, we consider the problem of selecting variables to include in transport estimators. We provide a brief overview of the transportability framework and illustrate that while selection diagrams are a vital first step in variable selection, these graphs alone identify a sufficient but not strictly necessary set of variables for generating an unbiased transport estimate. Next, we conduct a simulation experiment assessing the impact of including unnecessary variables on the performance of the parametric g-computation transport estimator. Our results highlight that the types of variables included can affect the bias, variance, and mean squared error of the estimates. We find that addition of variables that are not causes of the outcome but whose distributions differ between the source and target populations can increase the variance and mean squared error of the transported estimates. On the other hand, inclusion of variables that are causes of the outcome (regardless of whether they modify the causal contrast of interest or differ in distribution between the populations) reduces the variance of the estimates without increasing the bias. Finally, exclusion of variables that cause the outcome but do not modify the causal contrast of interest does not increase bias. These findings suggest that variable selection approaches for transport should prioritize identifying and including all causes of the outcome in the study population rather than focusing on variables whose distribution may differ between the study sample and target population., Comment: Under Review

Pharmaceutical companies continue to seek innovative ways to explore whether a drug under development is likely to be suitable for all or only an identifiable stratum of patients in the target population. The sooner this can be done during the clinical development process, the better it is for the company, and downstream for prescribers, payers, and most importantly, for patients. To help enable this vision of stratified medicine, we describe a powerful statistical framework, Patient Response Identifiers for Stratified Medicine (PRISM), for the discovery of potential predictors of drug response and associated subgroups using machine learning tools. PRISM is highly flexible and can have many "configurations", allowing the incorporation of complementary models or tools for a variety of outcomes and settings. One promising PRISM configuration is to use the observed outcomes for subgroup identification, while using counterfactual within-patient predicted treatment differences for subgroup-specific treatment estimates and associated interpretation. This separates the "subgroup-identification" from the "decision-making" and, to facilitate clinical design planning, is a simple way to obtain unbiased treatment effect sizes in the discovered subgroups. Simulation results, along with data from a real clinical trial are used to illustrate the utility of the proposed PRISM framework., Comment: 24 pages; 6 figures, 3 tables