
Fisher AutoEncoders
It has been conjectured that the Fisher divergence is more robust to mod...
read it

VarNet: Variational Neural Networks for the Solution of Partial Differential Equations
In this paper we propose a new modelbased unsupervised learning method,...
read it

A Meanfield Analysis of Deep ResNet and Beyond: Towards Provable Optimization Via Overparameterization From Depth
Training deep neural networks with stochastic gradient descent (SGD) can...
read it

Mitigating Manipulation in Peer Review via Randomized Reviewer Assignments
We consider three important challenges in conference peer review: (i) re...
read it

Toward Automatic Threat Recognition for Airport Xray Baggage Screening with Deep Convolutional Object Detection
For the safety of the traveling public, the Transportation Security Admi...
read it

ParkingSticker: A RealWorld Object Detection Dataset
We present a new and challenging object detection dataset, ParkingSticke...
read it

Mapping Motor Cortex Stimulation to Muscle Responses: A Deep Neural Network Modeling Approach
A deep neural network (DNN) that can reliably model muscle responses fro...
read it

Generative Adversarial Network Training is a Continual Learning Problem
Generative Adversarial Networks (GANs) have proven to be a powerful fram...
read it

DeepObfuscator: Adversarial Training Framework for PrivacyPreserving Image Classification
Deep learning has been widely utilized in many computer vision applicati...
read it

Enhancing Crosstask BlackBox Transferability of Adversarial Examples with Dispersion Reduction
Neural networks are known to be vulnerable to carefully crafted adversar...
read it

Stochastic Gradient Descent Escapes Saddle Points Efficiently
This paper considers the perturbed stochastic gradient descent algorithm...
read it

How You Act Tells a Lot: PrivacyLeakage Attack on Deep Reinforcement Learning
Machine learning has been widely applied to various applications, some o...
read it

Towards Understanding Fast Adversarial Training
Current neuralnetworkbased classifiers are susceptible to adversarial ...
read it

A Short Note on Concentration Inequalities for Random Vectors with SubGaussian Norm
In this note, we derive concentration inequalities for random vectors wi...
read it

In Pursuit of Interpretable, Fair and Accurate Machine Learning for Criminal Recidivism Prediction
In recent years, academics and investigative journalists have criticized...
read it

LaneNet: RealTime Lane Detection Networks for Autonomous Driving
Lane detection is to detect lanes on the road and provide the accurate l...
read it

Adversarial Defense via Data Dependent Activation Function and Total Variation Minimization
We improve the robustness of deep neural nets to adversarial attacks by ...
read it

Nonparametric Deconvolution Models
We describe nonparametric deconvolution models (NDMs), a family of Bayes...
read it

ModelBased Learning of Turbulent Flows using Mobile Robots
In this paper we consider the problem of modelbased learning of turbule...
read it

The Step Decay Schedule: A Near Optimal, Geometrically Decaying Learning Rate Procedure
There is a stark disparity between the step size schedules used in pract...
read it

A DeepLearning Algorithm for Thyroid Malignancy Prediction From Whole Slide Cytopathology Images
We consider thyroidmalignancy prediction from ultrahighresolution who...
read it

LotteryFL: Personalized and CommunicationEfficient Federated Learning with Lottery Ticket Hypothesis on NonIID Datasets
Federated learning is a popular distributed machine learning paradigm wi...
read it

Learning Diverse Fashion Collocation by Neural Graph Filtering
Fashion recommendation systems are highly desired by customers to find v...
read it

DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensembles
Recent research finds CNN models for image classification demonstrate ov...
read it

Survival Function Matching for Calibrated TimetoEvent Predictions
Models for predicting the time of a future event are crucial for risk as...
read it

Influence Function based Data Poisoning Attacks to TopN Recommender Systems
Recommender system is an essential component of web services to engage u...
read it

Learning Lowrank Deep Neural Networks via Singular Vector Orthogonality Regularization and Singular Value Sparsification
Modern deep neural networks (DNNs) often require high memory consumption...
read it

SpiderBoost: A Class of Faster Variancereduced Algorithms for Nonconvex Optimization
There has been extensive research on developing stochastic variance redu...
read it

Learning Diverse Stochastic HumanAction Generators by Learning Smooth Latent Transitions
Humanmotion generation is a longstanding challenging task due to the r...
read it

Geometric Matrix Completion with Deep Conditional Random Fields
The problem of completing highdimensional matrices from a limited set o...
read it

Knowledgedriven Encode, Retrieve, Paraphrase for Medical Image Report Generation
Generating long and semanticcoherent reports to describe medical images...
read it

TRP: Trained Rank Pruning for Efficient Deep Neural Networks
To enable DNNs on edge devices like mobile phones, lowrank approximatio...
read it

Improved Design of Quadratic Discriminant Analysis Classifier in Unbalanced Settings
The use of quadratic discriminant analysis (QDA) or its regularized vers...
read it

Low to High Dimensional Modality Hallucination using Aggregated Fields of View
Realworld robotics systems deal with data from a multitude of modalitie...
read it

NTIRE 2020 Challenge on Image and Video Deblurring
Motion blur is one of the most common degradation artifacts in dynamic s...
read it

Cooperative MultiAgent Reinforcement Learning with Partial Observations
In this paper, we propose a distributed zerothorder policy optimization...
read it

This looks like that: deep learning for interpretable image recognition
When we are faced with challenging image classification tasks, we often ...
read it

Expression of Fractals Through Neural Network Functions
To help understand the underlying mechanisms of neural networks (NNs), s...
read it

Detecting Adversarial Samples Using Influence Functions and Nearest Neighbors
Deep neural networks (DNNs) are notorious for their vulnerability to adv...
read it

Discretized Bottleneck in VAE: PosteriorCollapseFree SequencetoSequence Learning
Variational autoencoders (VAEs) are important tools in endtoend repres...
read it

StraightThrough Estimator as Projected Wasserstein Gradient Flow
The StraightThrough (ST) estimator is a widely used technique for back...
read it

POINTER: Constrained Text Generation via Insertionbased Generative Pretraining
Largescale pretrained language models, such as BERT and GPT2, have ac...
read it

Improving Disentangled Text Representation Learning with InformationTheoretic Guidance
Learning disentangled representations of natural language is essential f...
read it

Thyroid Cancer Malignancy Prediction From Whole Slide Cytopathology Images
We consider preoperative prediction of thyroid cancer based on ultrahig...
read it

Explaining Landscape Connectivity of Lowcost Solutions for Multilayer Nets
Mode connectivity is a surprising phenomenon in the loss landscape of de...
read it

TIPRDC: TaskIndependent PrivacyRespecting Data Crowdsourcing Framework with Anonymized Intermediate Representations
The success of deep learning partially benefits from the availability of...
read it

MVStylizer: An Efficient EdgeAssisted Video Photorealistic Style Transfer System for Mobile Phones
Recent research has made great progress in realizing neural style transf...
read it

AdvDetPatch: Attacking Object Detectors with Adversarial Patches
Object detectors have witnessed great progress in recent years and have ...
read it

Solving Jigsaw Puzzles By The Graph Connection Laplacian
We propose a novel mathematical framework to address the problem of auto...
read it

Scalable Thompson Sampling via Optimal Transport
Thompson sampling (TS) is a class of algorithms for sequential decision...
read it
Duke University
Duke's Office of News and Communications and Duke Web Services (formerly Blackwell Interactive) took the lead in developing the current version, in close collaboration with other campus groups. Viget Labs, a local design firm, assisted with part of the project. The web team has developed a set of free tools to help other members of the Duke community create or modify their sites to reflect the "look and feel" of the university's home page. These tools include messaging and style guidelines, color palettes, typefaces, Web templates and information about using the "Duke brand bar." For more information, please consult the online style guide or contact Duke Web Services.