arXiv subjects

Bin Yu

Publications and source records attributed to Bin Yu.

At least 145 records · Page 8Linked to original sources

arXiv2020

The geometric realization of regular path complexes via (co-)homology

The aim of this paper is to give the geometric realization of regular path complexes via (co)homology groups with coefficients in a ring $R$. Concretely, for each regular path complex $P$, we associate it with a singular $Δ$-complex $S(P)$ and show that the (co)homology groups of $P$ are isomorphic to those of $S(P)$ with coefficients in $R$. As a direct result we recognize path (co)homology as Hochschild (co)homology in case that $R$ is commutative and $P$ regular finite. Analogues of the Eilenberg-Zilber theorem and Künneth formula are also showed for the Cartesian product and the join of two regular path complexes. In fact, we meanwhile improve some previous results which are covered by these conclusions in this paper.

math.RT↗

arXiv2020

A stability-driven protocol for drug response interpretable prediction (staDRIP)

Modern cancer -omics and pharmacological data hold great promise in precision cancer medicine for developing individualized patient treatments. However, high heterogeneity and noise in such data pose challenges for predicting the response of cancer cell lines to therapeutic drugs accurately. As a result, arbitrary human judgment calls are rampant throughout the predictive modeling pipeline. In this work, we develop a transparent stability-driven pipeline for drug response interpretable predictions, or staDRIP, which builds upon the PCS framework for veridical data science (Yu and Kumbier, 2020) and mitigates the impact of human judgment calls. Here we use the PCS framework for the first time in cancer research to extract proteins and genes that are important in predicting the drug responses and stable across appropriate data and model perturbations. Out of the 24 most stable proteins we identified using data from the Cancer Cell Line Encyclopedia (CCLE), 18 have been associated with the drug response or identified as a known or possible drug target in previous literature, demonstrating the utility of our stability-driven pipeline for knowledge discovery in cancer drug response prediction modeling.

q-bio.QM↗

arXiv2020

Interpretations are useful: penalizing explanations to align neural networks with prior knowledge

For an explanation of a deep learning model to be effective, it must provide both insight into a model and suggest a corresponding action in order to achieve some objective. Too often, the litany of proposed explainable deep learning methods stop at the first step, providing practitioners with insight into a model, but no way to act on it. In this paper, we propose contextual decomposition explanation penalization (CDEP), a method which enables practitioners to leverage existing explanation methods in order to increase the predictive accuracy of deep learning models. In particular, when shown that a model has incorrectly assigned importance to some features, CDEP enables practitioners to correct these errors by directly regularizing the provided explanations. Using explanations provided by contextual decomposition (CD) (Murdoch et al., 2018), we demonstrate the ability of our method to increase performance on an array of toy and real datasets.

cs.LG↗

arXiv2020

Stable discovery of interpretable subgroups via calibration in causal studies

Building on Yu and Kumbier's PCS framework and for randomized experiments, we introduce a novel methodology for Stable Discovery of Interpretable Subgroups via Calibration (StaDISC), with large heterogeneous treatment effects. StaDISC was developed during our re-analysis of the 1999-2000 VIGOR study, an 8076 patient randomized controlled trial (RCT), that compared the risk of adverse events from a then newly approved drug, Rofecoxib (Vioxx), to that from an older drug Naproxen. Vioxx was found to, on average and in comparison to Naproxen, reduce the risk of gastrointestinal (GI) events but increase the risk of thrombotic cardiovascular (CVT) events. Applying StaDISC, we fit 18 popular conditional average treatment effect (CATE) estimators for both outcomes and use calibration to demonstrate their poor global performance. However, they are locally well-calibrated and stable, enabling the identification of patient groups with larger than (estimated) average treatment effects. In fact, StaDISC discovers three clinically interpretable subgroups each for the GI outcome (totaling 29.4% of the study size) and the CVT outcome (totaling 11.0%). Complementary analyses of the found subgroups using the 2001-2004 APPROVe study, a separate independently conducted RCT with 2587 patients, provides further supporting evidence for the promise of StaDISC.

stat.ME↗

arXiv2020

Fast Kernelized Correlation Filters without Boundary Effect

In recent years, correlation filter based trackers (CF trackers) have attracted much attention from the vision community because of their top performance in both localization accuracy and efficiency. The society of visual tracking, however, still needs to deal with the following difficulty on CF trackers: avoiding or eliminating the boundary effect completely, in the meantime, exploiting non-linear kernels and running efficiently. In this paper, we propose a fast kernelized correlation filter without boundary effect (nBEKCF) to solve this problem. To avoid the boundary effect thoroughly, a set of \emph{real} and \emph{dense} patches is sampled through the traditional sliding window and used as the training samples to train nBEKCF to fit a Gaussian response map. Non-linear kernels can be applied naturally in nBEKCF due to its different theoretical foundation from the existing CF trackers'. To achieve the fast training and detection, a set of cyclic bases is introduced to construct the filter. Two algorithms, ACSII and CCIM, are developed to significantly accelerate the calculation of kernel correlation matrices. ACSII and CCIM fully exploit the density of training samples and cyclic structure of bases, and totally run in space domain. The efficiency of CCIM exceeds that of the FFT counterpart remarkably in our task. Extensive experiments on six public datasets, OTB-2013, OTB-2015, NfS, VOT2018, GOT10k, and TrackingNet, show that compared to the CF trackers designed to relax the boundary effect, BACF and SRDCF, our nBEKCF achieves higher localization accuracy without tricks, in the meanwhile, runs at higher FPS.

cs.CV↗

arXiv2020

Curating a COVID-19 data repository and forecasting county-level death counts in the United States

As the COVID-19 outbreak evolves, accurate forecasting continues to play an extremely important role in informing policy decisions. In this paper, we present our continuous curation of a large data repository containing COVID-19 information from a range of sources. We use this data to develop predictions and corresponding prediction intervals for the short-term trajectory of COVID-19 cumulative death counts at the county-level in the United States up to two weeks ahead. Using data from January 22 to June 20, 2020, we develop and combine multiple forecasts using ensembling techniques, resulting in an ensemble we refer to as Combined Linear and Exponential Predictors (CLEP). Our individual predictors include county-specific exponential and linear predictors, a shared exponential predictor that pools data together across counties, an expanded shared exponential predictor that uses data from neighboring counties, and a demographics-based shared exponential predictor. We use prediction errors from the past five days to assess the uncertainty of our death predictions, resulting in generally-applicable prediction intervals, Maximum (absolute) Error Prediction Intervals (MEPI). MEPI achieves a coverage rate of more than 94% when averaged across counties for predicting cumulative recorded death counts two weeks in the future. Our forecasts are currently being used by the non-profit organization, Response4Life, to determine the medical supply need for individual hospitals and have directly contributed to the distribution of medical supplies across the country. We hope that our forecasts and data repository at https://covidseverity.com can help guide necessary county-specific decision-making and help counties prepare for their continued fight against COVID-19.

stat.AP↗

arXiv2020

Incremental causal effects

Causal evidence is needed to act and it is often enough for the evidence to point towards a direction of the effect of an action. For example, policymakers might be interested in estimating the effect of slightly increasing taxes on private spending across the whole population. We study identifiability and estimation of causal effects, where a continuous treatment is slightly shifted across the whole population (termed average partial effect or incremental causal effect). We show that incremental effects are identified under local ignorability and local overlap assumptions, where exchangeability and positivity only hold in a neighborhood of units. Average treatment effects are not identified under these assumptions. In this case, and under a smoothness condition, the incremental effect can be estimated via the average derivative. Moreover, we prove that in certain finite-sample observational settings, estimating the incremental effect is easier than estimating the average treatment effect in terms of asymptotic variance. For high-dimensional settings, we develop a simple feature transformation that allows for doubly-robust estimation and inference of incremental causal effects. Finally, we compare the behaviour of estimators of the incremental treatment effect and average treatment effect in experiments including data-inspired simulations.

stat.ME↗

arXiv2020

Two-stage growth mode for lift-off mechanism in oblique shock-wave/jet interaction

The lift-off characteristics of supersonic streamwise vortex in oblique shock-wave/jet interaction (OS/JI for short), extracted from a wall-mount ramp injector in scramjet, is studied through Large-eddy simulation method. Shocked helium jet is deformed into a pair of streamwise vortex with a co-rotating companion vortex showing the lift-off characteristic immediately after shock. Based on the objective coordinate system in frame of oblique shock structure, it is found that the nature of three-dimensional lift-off structure of a shockinduced streamwise vortex is inherently and precisely controlled by a two-stage growth mode of structure kinetics of a shock bubble interaction (SBI for short). The striking similar structures between OS/JI and SBI support the proposition that the lift-off of streamwise vortex is the result of a underlying two-dimensional vortical motion. By considering the first stage impulsive linear growth rate, an improved vortex propagation model suitable for SBI is proposed and validated. The lift-off phenomena of both numerical OS/JI case in this paper and wall-mounted ramp injector cases in literature are well explained under the two-stage structure kinetics model of SBI. This model further predicts that for higher free stream Mach number (M > 5), increasing ramp compression shows little effect on elevating streamwise vortex while evident lift-off may occur for lower Mach number (M < 3.5), which offers the new way for preliminary design of streamwise vortex-based ramp injector in scramjet.

physics.flu-dyn↗

arXiv2020

A Survey on Dynamic Network Embedding

Real-world networks are composed of diverse interacting and evolving entities, while most of existing researches simply characterize them as particular static networks, without consideration of the evolution trend in dynamic networks. Recently, significant progresses in tracking the properties of dynamic networks have been made, which exploit changes of entities and links in the network to devise network embedding techniques. Compared to widely proposed static network embedding methods, dynamic network embedding endeavors to encode nodes as low-dimensional dense representations that effectively preserve the network structures and the temporal dynamics, which is beneficial to multifarious downstream machine learning tasks. In this paper, we conduct a systematical survey on dynamic network embedding. In specific, basic concepts of dynamic network embedding are described, notably, we propose a novel taxonomy of existing dynamic network embedding techniques for the first time, including matrix factorization based, Skip-Gram based, autoencoder based, neural networks based and other embedding methods. Additionally, we carefully summarize the commonly used datasets and a wide variety of subsequent tasks that dynamic network embedding can benefit. Afterwards and primarily, we suggest several challenges that the existing algorithms faced and outline possible directions to facilitate the future research, such as dynamic embedding models, large-scale dynamic networks, heterogeneous dynamic networks, dynamic attributed networks, task-oriented dynamic network embedding and more embedding spaces.

cs.SI↗

arXiv2020

Singularity, Misspecification, and the Convergence Rate of EM

A line of recent work has analyzed the behavior of the Expectation-Maximization (EM) algorithm in the well-specified setting, in which the population likelihood is locally strongly concave around its maximizing argument. Examples include suitably separated Gaussian mixture models and mixtures of linear regressions. We consider over-specified settings in which the number of fitted components is larger than the number of components in the true distribution. Such misspecified settings can lead to singularity in the Fisher information matrix, and moreover, the maximum likelihood estimator based on $n$ i.i.d. samples in $d$ dimensions can have a non-standard $\mathcal{O}((d/n)^{\frac{1}{4}})$ rate of convergence. Focusing on the simple setting of two-component mixtures fit to a $d$-dimensional Gaussian distribution, we study the behavior of the EM algorithm both when the mixture weights are different (unbalanced case), and are equal (balanced case). Our analysis reveals a sharp distinction between these two cases: in the former, the EM algorithm converges geometrically to a point at Euclidean distance of $\mathcal{O}((d/n)^{\frac{1}{2}})$ from the true parameter, whereas in the latter case, the convergence rate is exponentially slower, and the fixed point has a much lower $\mathcal{O}((d/n)^{\frac{1}{4}})$ accuracy. Analysis of this singular case requires the introduction of some novel techniques: in particular, we make use of a careful form of localization in the associated empirical process, and develop a recursive argument to progressively sharpen the statistical rate.

math.ST↗

arXiv2020

Classifying expanding attractors on figure eight knot complement space and non-transitive Anosov flows on Franks-Williams manifold

The path closure of figure eight knot complement space, $N_0$, supports a natural DA (derived from Anosov) expanding attractor. Using this attractor, Franks-Williams constructed the first example of non-transitive Anosov flow on the manifold $M_0$ obtained by gluing two copies of $N_0$ through identity map along their boundaries, named by Franks-Williams manifold. In this paper, our main goal is to classify expanding attractors on $N_0$ and non-transitive Anosov flows on $M_0$. We prove that, up to orbit-equivalence, the DA expanding attractor is the unique expanding attractor supported by $N_0$, and the non-transitive Anosov flow constructed by Franks and Williams is the unique non-transitive Anosov flow admitted by $M_0$. Moreover, more general cases are also discussed. In particular, we completely classify non-transitive Anosov flows on a family of infinitely many toroidal $3$-manifolds with two hyperbolic pieces, obtained by gluing two copies of $N_0$ through any gluing homeomorphism.

math.DS↗

arXiv2020

Structural Compression of Convolutional Neural Networks

Deep convolutional neural networks (CNNs) have been successful in many tasks in machine vision, however, millions of weights in the form of thousands of convolutional filters in CNNs makes them difficult for human intepretation or understanding in science. In this article, we introduce CAR, a greedy structural compression scheme to obtain smaller and more interpretable CNNs, while achieving close to original accuracy. The compression is based on pruning filters with the least contribution to the classification accuracy. We demonstrate the interpretability of CAR-compressed CNNs by showing that our algorithm prunes filters with visually redundant functionalities such as color filters. These compressed networks are easier to interpret because they retain the filter diversity of uncompressed networks with order of magnitude less filters. Finally, a variant of CAR is introduced to quantify the importance of each image category to each CNN filter. Specifically, the most and the least important class labels are shown to be meaningful interpretations of each filter.

cs.CV↗

arXiv2020

A Systematic Study of the dust of Galactic Supernova Remnants I. The Distance and the Extinction

By combining the photometric, spectroscopic, and astrometric information of the stars in the sightline of SNRs, the distances to and the extinctions of 32 Galactic supernova remnants (SNRs) are investigated. The stellar atmospheric parameters are from the SDSS$-$DR14$/$APOGEE and LAMOST$-$DR5$/$LEGUE spectroscopic surveys. The multi-band photometry, from optical to infrared, are collected from the {\it Gaia}, APASS, Pan--STARRS1, 2MASS, and {\it WISE} surveys. With the calibrated {\it Gaia} distances of individual stars, the distances to 15 of 32 SNRs are well determined from their produced extinction and association with molecular clouds. The upper limits of distance are derived for 3 SNRs. The color excess ratios $E(g_{\rm P1}-λ) / E(g_{\rm P1}-r_{\rm P1})$ of 32 SNRs are calculated, and their variation with wavebands is fitted by a simple dust model. The inferred dust grain size distribution bifurcates: while the graphite grains have comparable size to the average ISM dust, the silicate grains are generally larger. Along the way, the average extinction law from optical to near-infrared of the Milky Way is derived from the 1.3 million star sample and found to agree with the CCM89 law with $R_{\rm V}=3.15$.

astro-ph.SR↗

arXiv2019

Log-concave sampling: Metropolis-Hastings algorithms are fast

We consider the problem of sampling from a strongly log-concave density in $\mathbb{R}^d$, and prove a non-asymptotic upper bound on the mixing time of the Metropolis-adjusted Langevin algorithm (MALA). The method draws samples by simulating a Markov chain obtained from the discretization of an appropriate Langevin diffusion, combined with an accept-reject step. Relative to known guarantees for the unadjusted Langevin algorithm (ULA), our bounds show that the use of an accept-reject step in MALA leads to an exponentially improved dependence on the error-tolerance. Concretely, in order to obtain samples with TV error at most $δ$ for a density with condition number $κ$, we show that MALA requires $\mathcal{O} \big(κd \log(1/δ) \big)$ steps, as compared to the $\mathcal{O} \big(κ^2 d/δ^2 \big)$ steps established in past work on ULA. We also demonstrate the gains of MALA over ULA for weakly log-concave densities. Furthermore, we derive mixing time bounds for the Metropolized random walk (MRW) and obtain $\mathcal{O}(κ)$ mixing time slower than MALA. We provide numerical examples that support our theoretical findings, and demonstrate the benefits of Metropolis-Hastings adjustment for Langevin-type sampling algorithms.

stat.ML↗

arXiv2019

Veridical Data Science

Building and expanding on principles of statistics, machine learning, and scientific inquiry, we propose the predictability, computability, and stability (PCS) framework for veridical data science. Our framework, comprised of both a workflow and documentation, aims to provide responsible, reliable, reproducible, and transparent results across the entire data science life cycle. The PCS workflow uses predictability as a reality check and considers the importance of computation in data collection/storage and algorithm design. It augments predictability and computability with an overarching stability principle for the data science life cycle. Stability expands on statistical uncertainty considerations to assess how human judgment calls impact data results through data and model/algorithm perturbations. Moreover, we develop inference procedures that build on PCS, namely PCS perturbation intervals and PCS hypothesis testing, to investigate the stability of data results relative to problem formulation, data cleaning, modeling decisions, and interpretations. We illustrate PCS inference through neuroscience and genomics projects of our own and others and compare it to existing methods in high dimensional, sparse linear model simulations. Over a wide range of misspecified simulation models, PCS inference demonstrates favorable performance in terms of ROC curves. Finally, we propose PCS documentation based on R Markdown or Jupyter Notebook, with publicly available, reproducible codes and narratives to back up human choices made throughout an analysis. The PCS workflow and documentation are demonstrated in a genomics case study available on Zenodo.

stat.ML↗

arXiv2019

A Debiased MDI Feature Importance Measure for Random Forests

Tree ensembles such as Random Forests have achieved impressive empirical success across a wide variety of applications. To understand how these models make predictions, people routinely turn to feature importance measures calculated from tree ensembles. It has long been known that Mean Decrease Impurity (MDI), one of the most widely used measures of feature importance, incorrectly assigns high importance to noisy features, leading to systematic bias in feature selection. In this paper, we address the feature selection bias of MDI from both theoretical and methodological perspectives. Based on the original definition of MDI by Breiman et al. for a single tree, we derive a tight non-asymptotic bound on the expected bias of MDI importance of noisy features, showing that deep trees have higher (expected) feature selection bias than shallow ones. However, it is not clear how to reduce the bias of MDI using its existing analytical expression. We derive a new analytical expression for MDI, and based on this new expression, we are able to propose a debiased MDI feature importance measure using out-of-bag samples, called MDI-oob. For both the simulated data and a genomic ChIP dataset, MDI-oob achieves state-of-the-art performance in feature selection from Random Forests for both deep and shallow trees.

stat.ML↗

arXiv2019

Three dimensional dust mapping of 12 supernovae remnants in the Galactic anticentre

We present three dimensional (3D) dust mapping of 12 supernova remnants (SNRs) in the Galactic anti-center (Galactic longitude $l$ between 150\degr\ and 210\degr) based on a recent 3D interstellar extinction map. The dust distribution of the regions which cover the full extents in the radio continuum for the individual SNRs are discussed. Four SNRs show significant spatial coincidences between molecular clouds (MCs) revealed from the 3D extinction mapping and the corresponding radio features. The results confirm the interactions between these SNRs and their surrounding MCs. Based on these correlations, we provide new distance estimates of the four SNRs, G189.1+3.0 (IC443, $d=1729^{+116}_{-94} \rm \,pc$), G190.9-2.2 ($d=1036^{+17}_{-81} \rm \,pc$), G205.5+0.5 ($d=941^{+96}_{-94}$ or $1257^{+92}_{-101} \rm \,pc$) and G213.0-0.6 ($d=1146^{+79}_{-80} \rm \,pc$). In addition, we find indirect evidences of potential interactions between SNRs and MCs for three other SNRs. New distance constraints are also given for these three SNRs.

astro-ph.GA↗

arXiv2019

CharBot: A Simple and Effective Method for Evading DGA Classifiers

Domain generation algorithms (DGAs) are commonly leveraged by malware to create lists of domain names which can be used for command and control (C&C) purposes. Approaches based on machine learning have recently been developed to automatically detect generated domain names in real-time. In this work, we present a novel DGA called CharBot which is capable of producing large numbers of unregistered domain names that are not detected by state-of-the-art classifiers for real-time detection of DGAs, including the recently published methods FANCI (a random forest based on human-engineered features) and LSTM.MI (a deep learning approach). CharBot is very simple, effective and requires no knowledge of the targeted DGA classifiers. We show that retraining the classifiers on CharBot samples is not a viable defense strategy. We believe these findings show that DGA classifiers are inherently vulnerable to adversarial attacks if they rely only on the domain name string to make a decision. Designing a robust DGA classifier may, therefore, necessitate the use of additional information besides the domain name alone. To the best of our knowledge, CharBot is the simplest and most efficient black-box adversarial attack against DGA classifiers proposed to date.

cs.LG↗