Search

Causal inference methods for observational data are increasingly recognized as a valuable complement to randomized clinical trials (RCTs). They can, under strong assumptions, emulate RCTs or help refine their focus. Our approach to causal inference uses causal directed acyclic graphs (DAGs). We are motivated by a concern that many observational studies in medicine begin without a clear definition of their objectives, without awareness of the scientific potential, and without tools to identify the necessary in itinere adjustments. We present and illustrate methods that provide "midway insights" during study's course, identify meaningful causal questions within the study's reach and point to the necessary data base enhancements for these questions to be meaningfully tackled. The method hinges on concepts of identification and positivity. Concepts are illustrated through an analysis of data generated by patients with aneurysmal Subarachnoid Hemorrhage (aSAH) halfway through a study, focusing in particular on the consequences of external ventricular drain (EVD) in strata of the aSAH population. In addition, we propose a method for multicenter studies, to monitor the impact of changes in practice at an individual center's level, by leveraging principles of instrumental variable (IV) inference., Comment: 30 pages in 13pt font. 4 TIKZ figures. 9 PDF figures

The preservation of the corticospinal tract (CST) is key to good motor recovery after stroke. The gold standard method of assessing the CST with imaging is diffusion tensor tractography. However, this is not available for most intracerebral haemorrhage (ICH) patients. Non-contrast CT scans are routinely available in most ICH diagnostic pipelines, but delineating white matter from a CT scan is challenging. We utilise nnU-Net, trained on paired diagnostic CT scans and high-directional diffusion tractography maps, to segment the CST from diagnostic CT scans alone, and we show our model reproduces diffusion based tractography maps of the CST with a Dice similarity coefficient of 57%. Surgical haematoma evacuation is sometimes performed after ICH, but published clinical trials to date show that whilst surgery reduces mortality, there is no evidence of improved functional recovery. Restricting surgery to patients with an intact CST may reveal a subset of patients for whom haematoma evacuation improves functional outcome. We investigated the clinical utility of our model in the MISTIE III clinical trial dataset. We found that our model's CST integrity measure significantly predicted outcome after ICH in the acute and chronic time frames, therefore providing a prognostic marker for patients to whom advanced diffusion tensor imaging is unavailable. This will allow for future probing of subgroups who may benefit from surgery., Comment: Accepted to Miccai Switch Workshop

Modern cloud has turned data services into easily accessible commodities. With just a few clicks, users are now able to access a catalog of data processing systems for a wide range of tasks. However, the cloud brings in both complexity and opportunity. While cloud users can quickly start an application by using various data services, it can be difficult to configure and optimize these services to gain the most value from them. For cloud providers, managing every aspect of an ever-increasing set of data services, while meeting customer SLAs and minimizing operational cost is becoming more challenging. Cloud technology enables the collection of significant amounts of workload traces and system telemetry. With the progress in data science (DS) and machine learning (ML), it is feasible and desirable to utilize a data-driven, ML-based approach to automate various aspects of data services, resulting in the creation of autonomous data services. This paper presents our perspectives and insights on creating autonomous data services on Azure. It also covers the future endeavors we plan to undertake and unresolved issues that still need attention., Comment: SIGMOD Companion of the 2023 International Conference on Management of Data. 2023

Modern analytical workloads are highly heterogeneous and massively complex, making generic query optimizers untenable for many customers and scenarios. As a result, it is important to specialize these optimizers to instances of the workloads. In this paper, we continue a recent line of work in steering a query optimizer towards better plans for a given workload, and make major strides in pushing previous research ideas to production deployment. Along the way we solve several operational challenges including, making steering actions more manageable, keeping the costs of steering within budget, and avoiding unexpected performance regressions in production. Our resulting system, QQ-advisor, essentially externalizes the query planner to a massive offline pipeline for better exploration and specialization. We discuss various aspects of our design and show detailed results over production SCOPE workloads at Microsoft, where the system is currently enabled by default.

We propose a secure and lightweight key based challenge obfuscation for strong PUFs. Our architecture is designed to be resilient against learning attacks. Our obfuscation mechanism uses non-linear feedback shift registers (NLFSRs). Responses are directly provided to the user, without error correction or extra post-processing steps. We also discuss the cost of protecting our architecture against power analysis attacks with clock randomization, and Boolean masking. Security against learning attacks is assessed using avalanche criterion, and deep-neural network attacks. We designed a testchip in 65 nm CMOS. When compared to the baseline arbiter PUF implementation, the cost increase of our proposed architecture is 1.27x, and 2.2x when using clock randomization, and Boolean masking, respectively.

We design, implement, and assess the security of several variations of the PUF-on-PUF (POP) architecture. We perform extensive experiments with deep neural networks (DNNs), showing results that endorse its resilience to learning attacks when using APUFs with 6, or more, stages in the first layer. Compositions using APUFs with 2, and 4 stages are shown vulnerable to DNN attacks. We reflect on such results, extending previous techniques of influential bits to assess stage bias in APUF instances. Our data shows that compositions not always preserve security properties of PUFs, the size of PUFs used plays a crucial role. We implemented a testchip in 65 nm CMOS to obtain accurate measurements of uniformity, uniqueness, and response stability for our POP implementations. Measurement results show that minimum bit error rate is obtained when using APUFs with 8 stages in the first layer, while fewer APUF stages lead to a large spread of bit error rate across different chips.

Strong physical unclonable functions (PUFs) provide a low-cost authentication primitive for resource constrained devices. However, most strong PUF architectures can be modeled through learning algorithms with a limited number of CRPs. In this paper, we introduce the concept of non-monotonic response quantization for strong PUFs. Responses depend not only on which path is faster, but also on the distance between the arriving signals. Our experiments show that the resulting PUF has increased security against learning attacks. To demonstrate, we designed and implemented a non-monotonically quantized ring-oscillator based PUF in 65 nm technology. Measurement results show nearly ideal uniformity and uniqueness, with bit error rate of 13.4% over the temperature range from 0 C to 50 C.

We discuss the prediction of the $W$ boson mass in a simple extension of the Standard Model ($\Sigma {\rm SM}$) with a real scalar triplet. A shift in the $W$ mass as reported by the CDF II collaboration can naturally be accommodated by the model without modifying the Standard Model value for the $Z$ mass. We discuss the main implications and the properties of the new Higgs bosons. Namely, the partial decay widths of the new charged Higgs are predicted. Furthermore, the neutral Higgs has suppressed couplings to fermions and decays predominantly into a pair of $W$ gauge bosons., Comment: 5 pages, 2 figures. v2: Added discussion on the scalar mixing angle, version to appear in Physics Letters B

Last-level cache (LLC) partitioning is a technique to provide temporal isolation and low worst-case latency (WCL) bounds when cores access the shared LLC in multicore safety-critical systems. A typical approach to cache partitioning involves allocating a separate partition to a distinct core. A central criticism of this approach is its poor utilization of cache storage. Today's trend of integrating a larger number of cores exacerbates this issue such that we are forced to consider shared LLC partitions for effective deployments. This work presents an approach to share LLC partitions among multiple cores while being able to provide low WCL bounds.

Easy-to-use programming interfaces paired with cloud-scale processing engines have enabled big data system users to author arbitrarily complex analytical jobs over massive volumes of data. However, as the complexity and scale of analytical jobs increase, they encounter a number of unforeseen problems, hotspots with large intermediate data on temporary storage, longer job recovery time after failures, and worse query optimizer estimates being examples of issues that we are facing at Microsoft. To address these issues, we propose Phoebe, an efficient learning-based checkpoint optimizer. Given a set of constraints and an objective function at compile-time, Phoebe is able to determine the decomposition of job plans, and the optimal set of checkpoints to preserve their outputs to durable global storage. Phoebe consists of three machine learning predictors and one optimization module. For each stage of a job, Phoebe makes accurate predictions for: (1) the execution time, (2) the output size, and (3) the start/end time taking into account the inter-stage dependencies. Using these predictions, we formulate checkpoint optimization as an integer programming problem and propose a scalable heuristic algorithm that meets the latency requirement of the production environment. We demonstrate the effectiveness of Phoebe in production workloads, and show that we can free the temporary storage on hotspots by more than 70% and restart failed jobs 68% faster on average with minimum performance impact. Phoebe also illustrates that adding multiple sets of checkpoints is not cost-efficient, which dramatically reduces the complexity of the optimization.

One class of solutions to the strong CP problem relies on generalized parity symmetries. A minimal model of this type, constructed by Babu and Mohapatra and based on a softly broken parity symmetry, has the remarkable property that effective QCD vacuum angle $\bar\theta$ vanishes up to one-loop order. We compute the leading two-loop contributions to $\bar\theta$ in this model and estimate subleading contributions. In contrast to previous estimates, we argue that $\bar \theta$ is not suppressed by the weak scale, and we find contributions of order $10^{-3}$-$10^{-2}$ multiplying unknown mixing angles and phases. Thus the model does not generically address the strong CP problem, but it might be made consistent with $\bar\theta<10^{-10}$ in some corners of parameter space. For such non-generic parameters, $\bar\theta$ is still likely to be just below present bounds, and therefore provides the dominant source of hadronic EDMs. We discuss the resulting EDM phenomenology., Comment: 8 pages, 1 figure

Optimizing resource allocation for analytical workloads is vital for reducing costs of cloud-data services. At the same time, it is incredibly hard for users to allocate resources per query in serverless processing systems, and they frequently misallocate by orders of magnitude. Unfortunately, prior work focused on predicting peak allocation while ignoring aggressive trade-offs between resource allocation and run-time. Additionally, these methods fail to predict allocation for queries that have not been observed in the past. In this paper, we tackle both these problems. We introduce a system for optimal resource allocation that can predict performance with aggressive trade-offs, for both new and past observed queries. We introduce the notion of a performance characteristic curve (PCC) as a parameterized representation that can compactly capture the relationship between resources and performance. To tackle training data sparsity, we introduce a novel data augmentation technique to efficiently synthesize the entire PCC using a single run of the query. Lastly, we demonstrate the advantages of a constrained loss function coupled with GNNs, over traditional ML methods, for capturing the domain specific behavior through an extensive experimental evaluation over SCOPE big data workloads at Microsoft.

Supermassive black hole binary mergers generate a stochastic gravitational wave background detectable by pulsar timing arrays. While the amplitude of this background is subject to significant uncertainties, the frequency dependence is a robust prediction of general relativity. We show that the effects of new forces beyond the Standard Model can modify this prediction and introduce unique features into the spectral shape. In particular, we consider the possibility that black holes in binaries are charged under a new long-range force, and we find that pulsar timing arrays are capable of robustly detecting such forces. Supermassive black holes and their environments can acquire charge due to high-energy particle production or dark sector interactions, making the measurement of the spectral shape a powerful test of fundamental physics., Comment: 10 pages, 3 figures. Matched published version

Axions have for some time been considered a plausible candidate for dark matter. They can be produced through misalignment, but it has been argued that when inflation occurs before a Peccei-Quinn transition, appreciable production can result from cosmic strings. This has been the subject of extensive simulations. But there are reasons to be skeptical about the possible role of axion strings. We review and elaborate on these questions, and argue that parametrically strings are already accounted for by the assumption of random misalignment angles. We review and elaborate on these questions, and provide several qualitative arguments that parametrically strings are already accounted for by the assumption of random misalignment angles. The arguments are base on considerations of the collective modes of the string solutions, on computations of axion radiation in particular models, and reviews of simulations., Comment: 31 pages, 3 figures; published version

A solution to the Boltzmann equation governing the thermal relic abundance of cold dark matter is constructed by matched asymptotic approximations. The approximation of the relic density is an asymptotic series valid when the abundance does not deviate significantly from its equilibrium value until small temperatures. Resonance and threshold effects are taken into account at leading order and found to be negligible unless the annihilation cross section is negligible at threshold. Comparisons are made to previously attempted constructions and to the freeze out approximation commonly employed in the literature. Extensions to higher order matching is outlined, and implications for solving related systems are discussed. We compare our results to a numerical determination of the relic abundance using a benchmark model and find a fantastic agreement. The method developed also serves as a solution to a wide class of problems containing an infinite order turning point., Comment: 15 pages, 6 figures

We present the first complete two loop calculation of the electron EDM in the complex two-Higgs doublet model. We confirm gauge-independence by demonstrating analytic cancellation of the gauge parameter $\xi$ in the background field gauge and the 't Hooft $R_\xi$ gauge. We also investigate the behavior of the electron EDM near the decoupling limit, and determine the short- and long-distance contributions by matching onto an effective field theory. Compared with earlier studies of the electron EDM in the complex two-Higgs doublet model, we note disagreements in several places and provide diagnoses where possible. We also provide expressions for EDMs of light quarks., Comment: 18 pages, 14 figures v2: references added; comparison with 1311.4704 corrected; Mathematica notebook containing all relevant equations added as ancillary file

It has been suggested that scattering cross sections at very high energies for producing large numbers of Higgs particles may exhibit factorial growth, and that curing this growth might be relevant to other questions in the Standard Model. We point out, first, that the question is inherently non-perturbative; low orders in the formal perturbative expansion do not give a good approximation to the scattering amplitude for sufficiently large N for any fixed, small value of the coupling. Focusing on $\lambda \phi^{4}$ theory, we argue that there may be a systematic approximation scheme for processes where N particles near threshold scatter to produce N particles, and discuss the leading contributions to the scattering amplitude and cross sections in this limit. Scattering amplitudes do not grow as rapidly as in perturbation theory. Additionally, partial and total cross sections do not show factorial growth. In the case of cross sections for $2 \to N$ particles, there is no systematic large N approximation available. That said, we provide evidence that non-perturbatively, there is no factorial growth in partial or total cross sections., Comment: 22 pages, 3 figures

Query processing over big data is ubiquitous in modern clouds, where the system takes care of picking both the physical query execution plans and the resources needed to run those plans, using a cost-based query optimizer. A good cost model, therefore, is akin to better resource efficiency and lower operational costs. Unfortunately, the production workloads at Microsoft show that costs are very complex to model for big data systems. In this work, we investigate two key questions: (i) can we learn accurate cost models for big data systems, and (ii) can we integrate the learned models within the query optimizer. To answer these, we make three core contributions. First, we exploit workload patterns to learn a large number of individual cost models and combine them to achieve high accuracy and coverage over a long period. Second, we propose extensions to Cascades framework to pick optimal resources, i.e, number of containers, during query planning. And third, we integrate the learned cost models within the Cascade-style query optimizer of SCOPE at Microsoft. We evaluate the resulting system, Cleo, in a production environment using both production and TPC-H workloads. Our results show that the learned cost models are 2 to 3 orders of magnitude more accurate, and 20X more correlated with the actual runtimes, with a large majority (70%) of the plan changes leading to substantial improvements in latency as well as resource usage., Comment: To appear at SIGMOD 2020

Vector leptoquarks can address the lepton flavor universality anomalies in decays associated with the $b \to c \ell \nu$ and $b \to s \ell \ell$ transitions, as observed in recent years. Generically, these leptoquarks yield new sources of CP violation. In this paper, we explore constraints and discovery potential for electric dipole moments (EDMs) in leptonic and hadronic systems. We provide the most generic expressions for dipole moments induced by vector leptoquarks at one loop. We find that $O(1)$ CP-violating phases in tau and muon couplings can lead to corresponding EDMs within reach of next-generation EDM experiments, and that existing bounds on the electron EDM already put stringent constraints on CP-violating electron couplings., Comment: 36 pages, 4 figures; muon EDM projection updated, references added, conclusions unchanged