Private Machine Learning in TensorFlow using Secure Computation

18 Oct 20182 code implementations

We present a framework for experimenting with secure multi-party computation directly in TensorFlow.

Very Deep Convolutional Networks for Large-Scale Image Recognition

4 Sep 2014122 code implementations

In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting.

IMAGE CLASSIFICATION

A generic framework for privacy preserving deep learning

9 Nov 20183 code implementations

We detail a new framework for privacy preserving deep learning and discuss its assets.

PRIVACY PRESERVING DEEP LEARNING

Active Anomaly Detection via Ensembles

17 Sep 20182 code implementations

First, we present an important insight into how anomaly detector ensembles are naturally suited for active learning.

ACTIVE LEARNING ANOMALY DETECTION

An IoT Endpoint System-on-Chip for Secure and Energy-Efficient Near-Sensor Analytics

18 Dec 20163 code implementations

Near-sensor data analytics is a promising direction for IoT endpoints, as it minimizes energy spent on communication and reduces network load - but it also poses security concerns, as valuable data is stored or sent over the network at various stages of the analytics pipeline.

EEG FACE DETECTION SEIZURE DETECTION

Active Anomaly Detection via Ensembles: Insights, Algorithms, and Interpretability

23 Jan 20192 code implementations

In this paper, we study the problem of active learning to automatically tune ensemble of anomaly detectors to maximize the number of true anomalies discovered.

ACTIVE LEARNING ANOMALY DETECTION

Defending Against Neural Fake News

NeurIPS 2019 2 code implementations

We find that best current discriminators can classify neural fake news from real, human-written, news with 73% accuracy, assuming access to a moderate level of training data.

FAKE NEWS DETECTION TEXT GENERATION

Universal adversarial perturbations

CVPR 2017 7 code implementations

Given a state-of-the-art deep neural network classifier, we show the existence of a universal (image-agnostic) and very small perturbation vector that causes natural images to be misclassified with high probability.

HydraPlus-Net: Attentive Deep Features for Pedestrian Analysis

ICCV 2017 1 code implementation

Pedestrian analysis plays a vital role in intelligent video surveillance and is a key component for security-centric computer vision systems.

PEDESTRIAN ATTRIBUTE RECOGNITION PERSON RE-IDENTIFICATION

Low Latency Privacy Preserving Inference

ICLR 2019 1 code implementation

When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity.

TRANSFER LEARNING