User profiles for Borja Balle
Borja BalleDeepMind Verified email at google.com Cited by 6331 |
Improving the gaussian mechanism for differential privacy: Analytical calibration and optimal denoising
The Gaussian mechanism is an essential building block used in multitude of differentially
private data analysis algorithms. In this paper we revisit the Gaussian mechanism and show …
private data analysis algorithms. In this paper we revisit the Gaussian mechanism and show …
The privacy blanket of the shuffle model
This work studies differential privacy in the context of the recently proposed shuffle model.
Unlike in the local model, where the server collecting privatized data from users can track back …
Unlike in the local model, where the server collecting privatized data from users can track back …
Unlocking high-accuracy differentially private image classification through scale
Differential Privacy (DP) provides a formal privacy guarantee preventing adversaries with
access to a machine learning model from extracting information about individual training points…
access to a machine learning model from extracting information about individual training points…
Subsampled rényi differential privacy and analytical moments accountant
We study the problem of subsampling in differential privacy (DP), a question that is the
centerpiece behind many successful differentially private machine learning algorithms. …
centerpiece behind many successful differentially private machine learning algorithms. …
Learning weighted automata
Weighted finite automata (WFA) are finite automata whose transitions and states are augmented
with some weights, elements of a semiring. A WFA induces a function over strings. The …
with some weights, elements of a semiring. A WFA induces a function over strings. The …
Privacy amplification by subsampling: Tight analyses via couplings and divergences
Differential privacy comes equipped with multiple analytical tools for the design of private
data analyses. One important tool is the so-called" privacy amplification by subsampling" …
data analyses. One important tool is the so-called" privacy amplification by subsampling" …
Extracting training data from diffusion models
Image diffusion models such as DALL-E 2, Imagen, and Stable Diffusion have attracted
significant attention due to their ability to generate high-quality synthetic images. In this work, we …
significant attention due to their ability to generate high-quality synthetic images. In this work, we …
Ethical and social risks of harm from language models
This paper aims to help structure the risk landscape associated with large-scale Language
Models (LMs). In order to foster advances in responsible innovation, an in-depth …
Models (LMs). In order to foster advances in responsible innovation, an in-depth …
Reconstructing training data with informed adversaries
Given access to a machine learning model, can an adversary reconstruct the model’s training
data? This work studies this question from the lens of a powerful informed adversary who …
data? This work studies this question from the lens of a powerful informed adversary who …
Taxonomy of risks posed by language models
Responsible innovation on large-scale Language Models (LMs) requires foresight into and
in-depth understanding of the risks these models may pose. This paper develops a …
in-depth understanding of the risks these models may pose. This paper develops a …