
Stable discovery of interpretable subgroups via calibration in causal studies
Building on Yu and Kumbier's PCS framework and for randomized experiment...
read it

Revisiting complexity and the biasvariance tradeoff
The recent success of highdimensional models, such as deep neural netwo...
read it

Instability, Computational Efficiency and Statistical Accuracy
Many statistical estimators are defined as the fixed point of a datadep...
read it

Curating a COVID19 data repository and forecasting countylevel death counts in the United States
As the COVID19 outbreak continues to evolve, accurate forecasting conti...
read it

Transformation Importance with Applications to Cosmology
Machine learning lies at the heart of new possibilities for scientific d...
read it

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 pro...
read it

Incremental causal effects
This is a draft. The ignorability assumption is a key assumption in caus...
read it

A Debiased MDI Feature Importance Measure for Random Forests
Tree ensembles such as Random Forests have achieved impressive empirical...
read it

Fast mixing of Metropolized Hamiltonian Monte Carlo: Benefits of multistep gradients
Hamiltonian Monte Carlo (HMC) is a stateoftheart Markov chain Monte C...
read it

Disentangled Attribution Curves for Interpreting Random Forests and Boosted Trees
Tree ensembles, such as random forests and AdaBoost, are ubiquitous mach...
read it

CharBot: A Simple and Effective Method for Evading DGA Classifiers
Domain generation algorithms (DGAs) are commonly leveraged by malware to...
read it

Unique Sharp Local Minimum in ℓ_1minimization Complete Dictionary Learning
We study the problem of globally recovering a dictionary from a set of s...
read it

Challenges with EM in application to weakly identifiable mixture models
We study a class of weakly identifiable locationscale mixture models fo...
read it

Three principles of data science: predictability, computability, and stability (PCS)
We propose the predictability, computability, and stability (PCS) framew...
read it

Interpretable machine learning: definitions, methods, and applications
Machinelearning models have demonstrated great success in learning comp...
read it

Mobile Robot Localisation and Navigation Using LEGO NXT and Ultrasonic Sensor
Mobile robots are becoming increasingly important both for individuals a...
read it

Refining interaction search through signed iterative Random Forests
Advances in supervised learning have enabled accurate prediction in biol...
read it

Singularity, Misspecification, and the Convergence Rate of EM
A line of recent work has characterized the behavior of the EM algorithm...
read it

Highspeed Tracking with Multikernel Correlation Filters
Correlation filter (CF) based trackers are currently ranked top in terms...
read it

High Speed Kernelized Correlation Filters without Boundary Effect
Recently, correlation filter based trackers (CF trackers) have attracted...
read it

Hierarchical interpretations for neural network predictions
Deep neural networks (DNNs) have achieved impressive predictive performa...
read it

Stability and Convergence Tradeoff of Iterative Optimization Algorithms
The overall performance or expected excess risk of an iterative machine ...
read it

Improved OpenCLbased Implementation of Social Field Pedestrian Model
Two aspects of improvements are proposed for the OpenCLbased implementa...
read it

Towards the Standardization of Nonorthogonal Multiple Access for Next Generation Wireless Networks
Nonorthogonal multiple access (NoMA) as an efficient way of radio resou...
read it

Beyond Word Importance: Contextual Decomposition to Extract Interactions from LSTMs
The driving force behind the recent success of LSTMs has been their abil...
read it

Logconcave sampling: MetropolisHastings algorithms are fast!
We consider the problem of sampling from a strongly logconcave density ...
read it

Artificial Intelligence and Statistics
Artificial intelligence (AI) is intrinsically datadriven. It calls for ...
read it

Interpreting Convolutional Neural Networks Through Compression
Convolutional neural networks (CNNs) achieve stateoftheart performanc...
read it

Fast MCMC sampling algorithms on polytopes
We propose and analyze two new MCMC sampling algorithms, the Vaidya walk...
read it

Iterative Random Forests to detect predictive and stable highorder interactions
Genomics has revolutionized biology, enabling the interrogation of whole...
read it

Structural Compression of Convolutional Neural Networks Based on Greedy Filter Pruning
Convolutional neural networks (CNNs) have stateoftheart performance o...
read it

Formulas for Counting the Sizes of Markov Equivalence Classes of Directed Acyclic Graphs
The sizes of Markov equivalence classes of directed acyclic graphs play ...
read it

Optimal Subsampling Approaches for Large Sample Linear Regression
A significant hurdle for analyzing large sample data is the lack of effe...
read it

Local identifiability of l_1minimization dictionary learning: a sufficient and almost necessary condition
We study the theoretical properties of learning a dictionary from N sign...
read it

Error Rate Bounds and Iterative Weighted Majority Voting for Crowdsourcing
Crowdsourcing has become an effective and popular tool for humanpowered...
read it

Statistical guarantees for the EM algorithm: From population to samplebased analysis
We develop a general framework for proving rigorous guarantees on the pe...
read it

The geometry of kernelized spectral clustering
Clustering of data sets is a standard problem in many areas of science a...
read it

Impact of regularization on Spectral Clustering
The performance of spectral clustering can be considerably improved via ...
read it

Error Rate Bounds in Crowdsourcing Models
Crowdsourcing is an effective tool for humanpowered computation on many...
read it

A Statistical Perspective on Algorithmic Leveraging
One popular method for dealing with largescale data sets is sampling. F...
read it

Early stopping and nonparametric regression: An optimal datadependent stopping rule
The strategy of early stopping is a regularization technique based on ch...
read it

Estimation Stability with Cross Validation (ESCV)
Crossvalidation (CV) is often used to select the regularization paramet...
read it

Supplement to "Reversible MCMC on Markov equivalence classes of sparse directed acyclic graphs"
This supplementary material includes three parts: some preliminary resul...
read it

Reversible MCMC on Markov equivalence classes of sparse directed acyclic graphs
Graphical models are popular statistical tools which are used to represe...
read it

Complexity Analysis of the Lasso Regularization Path
The regularization path of the Lasso can be shown to be piecewise linear...
read it

Supervised Feature Selection in Graphs with Path Coding Penalties and Network Flows
We consider supervised learning problems where the features are embedded...
read it

Coclustering for directed graphs: the Stochastic coBlockmodel and spectral algorithm DiSim
Directed graphs have asymmetric connections, yet the current graph clust...
read it

The Lasso under Heteroscedasticity
The performance of the Lasso is well understood under the assumptions of...
read it

Data spectroscopy: Eigenspaces of convolution operators and clustering
This paper focuses on obtaining clustering information about a distribut...
read it
Bin Yu
verfied profile
Bin Yu is an American statistician and data scientist. She is Chancellor's Professor of Statistics and EECS at UC Berkeley and also an Investigator at ChanZuckerberg Biohub in San Francisco. Her research focuses on practice, algorithm, and theory of statistical machine learning and causal inference. Her group is engaged in interdisciplinary research with scientists from genomics, neuroscience, and precision medicine.
In order to augment empirical evidence for decisionmaking, she and her group are investigating methods/algorithms (and associated statistical inference problems) such as dictionary learning, nonnegative matrix factorization (NMF), EM and deep learning (CNNs and LSTMs), and heterogeneous effect estimation in randomized experiments (Xlearner). Their recent algorithms include staNMF for unsupervised learning, iterative Random Forests (iRF) and signed iRF (siRF) for discovering predictive and stable highorder interactions in supervised learning, and contextual decomposition (CD) and aggregated contextual decomposition (ACD) for phrase or patch importance extraction from an LSTM or a CNN.
She is a member of the U.S. National Academy of Sciences and Fellow of the American Academy of Arts and Sciences. She was a Guggenheim Fellow in 2006, and the Tukey Memorial Lecturer of the Bernoulli Society in 2012. She was President of IMS (Institute of Mathematical Statistics) in 20132014 and the Rietz Lecturer of IMS in 2016. She received the E. L. Scott Award from COPSS (Committee of Presidents of Statistical Societies) in 2018.