BYBin Yu

Bin Yu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Bin Yu has a total of 60 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Towards Robust Waveform-Based Acoustic Models

    by Dino Oglic, Zoran Cvetkovic, Peter Sollich, Steve Renals, Bin Yu

    We study the problem of learning robust acoustic models in adverse environments, characterized by a significant mismatch between training and test conditions. This problem is of paramount importance for the deployment of speech recognition systems that need to perform well in unseen environments. First, we characterize data augmentation theoretically as an instance of vicinal risk minimization, which aims at improving risk estimates during training by replacing the delta functions that define the empirical density over the input space with an approximation of the marginal population density in the vicinity of the training samples. More specifically, we assume that local neighborhoods centered at training samples can be approximated using a mixture of Gaussians, and demonstrate theoretically that this can incorporate robust inductive bias into the learning process. We then specify the individual mixture components implicitly via data augmentation schemes, designed to address common sources of spurious correlations in acoustic models. To avoid potential confounding effects on robustness due to information loss, which has been associated with standard feature extraction techniques (e.g., fbank and mfcc features), we focus on the waveform-based setting. Our empirical results show that the approach can generalize to unseen noise conditions, with 150% relative improvement in out-of-distribution generalization compared to training using the standard risk minimization principle. Moreover, the results demonstrate competitive performance relative to models learned using a training sample designed to match the acoustic conditions characteristic of test utterances.


  • Open Access

    Veridical data science

    by Bin Yu, Karl Kumbier

    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, composed of both a workflow and documentation, aims to provide responsible, reliable, reproducible, and transparent results across the 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. Stability expands on statistical uncertainty considerations to assess how human judgment calls impact data results through data and model/algorithm perturbations. As part of the PCS workflow, we develop PCS inference procedures, 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. Moreover, we demonstrate its favorable performance over existing methods in terms of receiver operating characteristic (ROC) curves in high-dimensional, sparse linear model simulations, including a wide range of misspecified models. 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.


  • Top Cited Papers
    Open Access

    Metalearners for estimating heterogeneous treatment effects using machine learning

    by Sören R. Künzel, Jasjeet S. Sekhon, Peter J. Bickel, Bin Yu

    Significance: Estimating and analyzing heterogeneous treatment effects is timely, yet challenging. We introduce a unifying framework for many conditional average treatment effect estimators, and we propose a metalearner, the X-learner, which can adapt to structural properties, such as the smoothness and sparsity of the underlying treatment effect. We present its favorable properties, using theory and simulations. We apply it, using random forests, to two field experiments in political science, where it is shown to be easy to use and to produce results that are interpretable.


  • Open Access

    The DeepTune framework for modeling and characterizing neurons in visual cortex area V4

    by Reza Abbasi-Asl, Yuansi Chen, Adam Bloniarz, Michael Oliver, Ben D.B. Willmore, Jack L. Gallant, Bin Yu

    Deep neural network models have recently been shown to be effective in predicting single neuron responses in primate visual cortex areas V4. Despite their high predictive accuracy, these models are generally difficult to interpret. This limits their applicability in characterizing V4 neuron function. Here, we propose the DeepTune framework as a way to elicit interpretations of deep neural network-based models of single neurons in area V4. V4 is a midtier visual cortical area in the ventral visual pathway. Its functional role is not yet well understood. Using a dataset of recordings of 71 V4 neurons stimulated with thousands of static natural images, we build an ensemble of 18 neural network-based models per neuron that accurately predict its response given a stimulus image. To interpret and visualize these models, we use a stability criterion to form optimal stimuli (DeepTune images) by pooling the 18 models together. These DeepTune images not only confirm previous findings on the presence of diverse shape and texture tuning in area V4, but also provide rich, concrete and naturalistic characterization of receptive fields of individual V4 neurons. The population analysis of DeepTune images for 71 neurons reveals how different types of curvature tuning are distributed in V4. In addition, it also suggests strong suppressive tuning for nearly half of the V4 neurons. Though we focus exclusively on the area V4, the DeepTune framework could be applied more generally to enhance the understanding of other visual cortex areas.


  • Artificial intelligence and statistics

    by Bin Yu, Karl Kumbier

    Artificial intelligence (AI) is intrinsically data-driven. It calls for the application of statistical concepts through human-machine collaboration during the generation of data, the development of...


  • Open Access

    Iterative random forests to discover predictive and stable high-order interactions

    by Sumanta Basu, Karl Kumbier, James B. Brown, Bin Yu

    Genomics has revolutionized biology, enabling the interrogation of whole transcriptomes, genome-wide binding sites for proteins, and many other molecular processes. However, individual genomic assays measure elements that interact in vivo as components of larger molecular machines. Understanding how these high-order interactions drive gene expression presents a substantial statistical challenge. Building on random forests (RFs) and random intersection trees (RITs) and through extensive, biologically inspired simulations, we developed the iterative random forest algorithm (iRF). iRF trains a feature-weighted ensemble of decision trees to detect stable, high-order interactions with the same order of computational cost as the RF. We demonstrate the utility of iRF for high-order interaction discovery in two prediction problems: enhancer activity in the early Drosophila embryo and alternative splicing of primary transcripts in human-derived cell lines. In Drosophila, among the 20 pairwise transcription factor interactions iRF identifies as stable (returned in more than half of bootstrap replicates), 80% have been previously reported as physical interactions. Moreover, third-order interactions, e.g., between Zelda (Zld), Giant (Gt), and Twist (Twi), suggest high-order relationships that are candidates for follow-up experiments. In human-derived cells, iRF rediscovered a central role of H3K36me3 in chromatin-mediated splicing regulation and identified interesting fifth- and sixth-order interactions, indicative of multivalent nucleosomes with specific roles in splicing regulation. By decoupling the order of interactions from the computational cost of identification, iRF opens additional avenues of inquiry into the molecular mechanisms underlying genome biology.


  • Open Access

    Formulas for Counting the Sizes of Markov Equivalence Classes of Directed Acyclic Graphs

    by Yangbo He, Bin Yu

    The sizes of Markov equivalence classes of directed acyclic graphs play important roles in measuring the uncertainty and complexity in causal learning. A Markov equivalence class can be represented by an essential graph and its undirected subgraphs determine the size of the class. In this paper, we develop a method to derive the formulas for counting the sizes of Markov equivalence classes. We first introduce a new concept of core graph. The size of a Markov equivalence class of interest is a polynomial of the number of vertices given its core graph. Then, we discuss the recursive and explicit formula of the polynomial, and provide an algorithm to derive the size formula via symbolic computation for any given core graph. The proposed size formula derivation sheds light on the relationships between the size of a Markov equivalence class and its representation graph, and makes size counting efficient, even when the essential graphs contain non-sparse undirected subgraphs.

  • Top Cited Papers

    Co-clustering directed graphs to discover asymmetries and directional communities

    by Karl Rohe, Tai Qin, Bin Yu

    Significance: This paper adds to the continuing and long-running interest in networks and network clustering. For directed networks, we propose the di-sim algorithm to capture asymmetries of connections and discover directional clusters. We illustrate this algorithm with three data examples: the Enron email network, the hyperlinked blog network during the 2004 US presidential election, and the chemical connections among the neurons in Caenorhabditis elegans . We identify informative and bottleneck nodes in all three networks. In particular, for the third example, di-sim finds bottleneck nodes that create a feedforward structure among clusters of nodes.


  • Lasso adjustments of treatment effect estimates in randomized experiments

    by Adam Bloniarz, Hanzhong Liu, Cun-Hui Zhang, Jasjeet S. Sekhon, Bin Yu

    We provide a principled way for investigators to analyze randomized experiments when the number of covariates is large. Investigators often use linear multivariate regression to analyze randomized experiments instead of simply reporting the difference of means between treatment and control groups. Their aim is to reduce the variance of the estimated treatment effect by adjusting for covariates. If there are a large number of covariates relative to the number of observations, regression may perform poorly because of overfitting. In such cases, the least absolute shrinkage and selection operator (Lasso) may be helpful. We study the resulting Lasso-based treatment effect estimator under the Neyman–Rubin model of randomized experiments. We present theoretical conditions that guarantee that the estimator is more efficient than the simple difference-of-means estimator, and we provide a conservative estimator of the asymptotic variance, which can yield tighter confidence intervals than the difference-of-means estimator. Simulation and data examples show that Lasso-based adjustment can be advantageous even when the number of covariates is less than the number of observations. Specifically, a variant using Lasso for selection and ordinary least squares (OLS) for estimation performs particularly well, and it chooses a smoothing parameter based on combined performance of Lasso and OLS.


  • Stability-driven nonnegative matrix factorization to interpret spatial gene expression and build local gene networks

    by Siqi Wu, Antony Joseph, Ann S. Hammonds, Susan E. Celniker, Bin Yu, Erwin Frise

    Significance: Despite the abundance of spatial gene expression data, extracting meaningful information to reveal how genes interact remains a challenge. We developed staNMF, a method that combines a powerful unsupervised learning algorithm, nonnegative matrix factorization (NMF), with a new stability criterion that selects the size of the dictionary or the set of principal patterns (PP). We demonstrate that PP give rise to a novel and concise representation of theDrosophilaembryonic spatial expression patterns and they correspond to biologically meaningful regions of theDrosophilaembryo. Furthermore, this new representation was used to automatically predict manual annotations, categorize gene expression patterns, and reconstruct the local gap gene network with high accuracy.