Data analysis, Data, Machine learning, Neural network, Scientific method, Test set

Generalizing Randomized Smoothing for Pointwise-Certified Defenses to Data Poisoning Attacks

On Sep 25, 2020
@rsalakhu shared
Generalizing Randomized Smoothing for Pointwise-Certified Defenses to Data Poisoning Attacks – Machine Learning Blog | ML@CMU | Carnegie Mellon University https://t.co/j8Z8RyE2VQ
Open

Adversarial examples—targeted, human-imperceptible modifications to a test input that cause a deep network to fail catastrophically—have taken the machine learning community by storm, with a large body of literature dedicated to understanding and preventing this phenomenon (see these ...

blog.ml.cmu.edu
On Sep 25, 2020
@rsalakhu shared
Generalizing Randomized Smoothing for Pointwise-Certified Defenses to Data Poisoning Attacks – Machine Learning Blog | ML@CMU | Carnegie Mellon University https://t.co/j8Z8RyE2VQ
Open

Generalizing Randomized Smoothing for Pointwise-Certified Defenses to Data Poisoning Attacks

Generalizing Randomized Smoothing for Pointwise-Certified Defenses to Data Poisoning Attacks

Adversarial examples—targeted, human-imperceptible modifications to a test input that cause a deep network to fail catastrophically—have taken the machine learning community by storm, with ...

A Simple Baseline for Bayesian Uncertainty in Deep Learning

A Simple Baseline for Bayesian Uncertainty in Deep Learning

While we primarily focus on image classification, we show that SWAG can significantly im- prove test perplexities of LSTM networks on language modeling problems, and in Appendix 7 we also ...

Machine Learning Project Structure: Stages, Roles, and Tools

Machine Learning Project Structure: Stages, Roles, and Tools

Roles: data scientist Tools: spreadsheets, automated solutions (Weka, Trim, Trifacta Wrangler, RapidMiner), MLaaS (Google Cloud AI, Amazon Machine Learning, Azure Machine Learning) Data ...

H2O Framework for Machine Learning

H2O Framework for Machine Learning

H2O is a scalable and fast open-source platform for machine learning. We will apply it to perform classification tasks. The dataset we are using is the Bank M…

Panning for gold in your data stream

Panning for gold in your data stream

Analytics engines can be powerful tools when you have a mountain of data and you know what valuable content you want to get out of it. But what happens when you don’t know what you’re ...

Army scientists train machine learning models to wrangle dirty data

Army scientists train machine learning models to wrangle dirty data

Army researchers have developed a new approach for training machine learning models that can better withstand dirty and deceptive data. Models trained under this method have greatly ...

Microsoft Research Blog

Microsoft Research Blog

By combining adversarial training with a technique called randomized smoothing, which enlists math from the 19th century, Microsoft researchers created a method that establishes ...

Python Machine Learning Tutorial, Scikit-Learn: Wine Snob Edition

Python Machine Learning Tutorial, Scikit-Learn: Wine Snob Edition

Step-by-step Python machine learning tutorial for building a model from start to finish using Scikit-Learn. We'll have some fun and predict wine quality!

Towards Machine Learning Systems Design

Towards Machine Learning Systems Design

<p>Machine learning solutions, in particular those based on deep learning methods, form an underpinning for the modern artificial intelligence revolution that has dominated popular press ...