
-
Adversarial Semantic Collisions
We study semantic collisions: texts that are semantically unrelated but ...
read it
-
You Autocomplete Me: Poisoning Vulnerabilities in Neural Code Completion
Code autocompletion is an integral feature of modern code editors and ID...
read it
-
Information Leakage in Embedding Models
Embeddings are functions that map raw input data to low-dimensional vect...
read it
-
Generalized Zero-shot ICD Coding
The International Classification of Diseases (ICD) is a list of classifi...
read it
-
Membership Encoding for Deep Learning
Machine learning as a service (MLaaS), and algorithm marketplaces are on...
read it
-
Overlearning Reveals Sensitive Attributes
`Overlearning' means that a model trained for a seemingly simple objecti...
read it
-
The Natural Auditor: How To Tell If Someone Used Your Words To Train Their Model
To help enforce data-protection regulations such as GDPR and detect unau...
read it
-
Inference Attacks Against Collaborative Learning
Collaborative machine learning and related techniques such as distribute...
read it
-
Chiron: Privacy-preserving Machine Learning as a Service
Major cloud operators offer machine learning (ML) as a service, enabling...
read it
-
Fooling OCR Systems with Adversarial Text Images
We demonstrate that state-of-the-art optical character recognition (OCR)...
read it
-
Kernel Distillation for Gaussian Processes
Gaussian processes (GPs) are flexible models that can capture complex st...
read it
-
Membership Inference Attacks against Machine Learning Models
We quantitatively investigate how machine learning models leak informati...
read it
-
Learning Genomic Representations to Predict Clinical Outcomes in Cancer
Genomics are rapidly transforming medical practice and basic biomedical ...
read it