Search Results for author: Nikola Jovanović

Found 9 papers, 6 papers with code

Watermark Stealing in Large Language Models

no code implementations29 Feb 2024 Nikola Jovanović, Robin Staab, Martin Vechev

LLM watermarking has attracted attention as a promising way to detect AI-generated content, with some works suggesting that current schemes may already be fit for deployment.

From Principle to Practice: Vertical Data Minimization for Machine Learning

1 code implementation17 Nov 2023 Robin Staab, Nikola Jovanović, Mislav Balunović, Martin Vechev

We propose a novel vertical DM (vDM) workflow based on data generalization, which by design ensures that no full-resolution client data is collected during training and deployment of models, benefiting client privacy by reducing the attack surface in case of a breach.

Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning

1 code implementation5 Jun 2023 Kostadin Garov, Dimitar I. Dimitrov, Nikola Jovanović, Martin Vechev

Malicious server (MS) attacks have enabled the scaling of data stealing in federated learning to large batch sizes and secure aggregation, settings previously considered private.

Decoder Federated Learning

Private and Reliable Neural Network Inference

1 code implementation27 Oct 2022 Nikola Jovanović, Marc Fischer, Samuel Steffen, Martin Vechev

We employ these building blocks to enable privacy-preserving NN inference with robustness and fairness guarantees in a system called Phoenix.

Fairness Privacy Preserving

FARE: Provably Fair Representation Learning with Practical Certificates

1 code implementation13 Oct 2022 Nikola Jovanović, Mislav Balunović, Dimitar I. Dimitrov, Martin Vechev

To produce a practical certificate, we develop and apply a statistical procedure that computes a finite sample high-confidence upper bound on the unfairness of any downstream classifier trained on FARE embeddings.

Fairness Representation Learning

LAMP: Extracting Text from Gradients with Language Model Priors

2 code implementations17 Feb 2022 Mislav Balunović, Dimitar I. Dimitrov, Nikola Jovanović, Martin Vechev

Recent work shows that sensitive user data can be reconstructed from gradient updates, breaking the key privacy promise of federated learning.

Federated Learning Language Modelling

On the Paradox of Certified Training

no code implementations12 Feb 2021 Nikola Jovanović, Mislav Balunović, Maximilian Baader, Martin Vechev

Certified defenses based on convex relaxations are an established technique for training provably robust models.

Towards Sparse Hierarchical Graph Classifiers

1 code implementation3 Nov 2018 Cătălina Cangea, Petar Veličković, Nikola Jovanović, Thomas Kipf, Pietro Liò

Recent advances in representation learning on graphs, mainly leveraging graph convolutional networks, have brought a substantial improvement on many graph-based benchmark tasks.

General Classification Graph Classification +3

Cannot find the paper you are looking for? You can Submit a new open access paper.