arXiv subjects

Peter Horvath

Publications and source records attributed to Peter Horvath.

10 recordsLinked to original sources

arXiv2026

Mitosis Detection in the Wild: Multi-Tumor and Context-Aware Generalization in the MIDOG 2025 Challenge

Automated mitosis detection is a well-established task in computational pathology. While previous benchmarks focused on scanner-induced domain shift, clinical "real-world" application requires models to be robust across the vast variance to be expected in the histological landscape. The MItosis DOmain Generalization (MIDOG) 2025 challenge was designed to evaluate algorithmic performance across unprecedented biological and contextual diversity. We curated a test dataset of 365 cases, encompassing 12 distinct human, canine and feline tumor types, digitized across multiple scanning platforms. Moving beyond hand-selected hotspots, the challenge required detection also in random tissue areas (representative of the whole slide detection situation) and challenging areas (areas rich in hard negatives). In the second track, we introduced the classification of atypical mitotic figures (AMFs). There were 18 teams submitting to the detection track, with F1 scores ranging up to 0.740. In the AMF detection track, we had 21 submissions with balanced accuracy values up to 0.908. Our analysis reveals that while most models perform reliably in traditional hotspots, significant performance degradation occurs in challenging ROIs, where false positive rates tripled. Furthermore, performance varied significantly across the 12 tumor types, highlighting "blind spots" in current state-of-the-art architectures when encountering rare or highly pleomorphic malignancies. Moreover, we evaluated the effectiveness of ensembling and found a mean increases of 1.5 and 1.3 percentage points in F1 score and balanced accuracy, respectively. In contrast, TTA showed no relevant improvement. MIDOG 2025 demonstrates that "in the wild" mitosis detection remains a significant hurdle. The transition from hotspot-only evaluation to a multi-contextual framework provides a more realistic proxy for clinical reliability.

cs.CV↗

arXiv2026

Kraken: Higher-order EM Side-Channel Attacks on DNNs in Near and Far Field

The multi-million dollar investment required for modern machine learning (ML) has made large ML models a prime target for theft. In response, the field of model stealing has emerged. Attacks based on physical side-channel information have shown that DNN model extraction is feasible, even on CUDA Cores in a GPU. For the first time, our work demonstrates parameter extraction on the specialized GPU's Tensor Core units, most commonly used GPU units nowadays due to their superior performance, via near-field physical side-channel attacks. Previous work targeted only the general-purpose CUDA Cores in the GPU, the functional units that have been part of the GPU since its inception. Our method is tailored to the GPU architecture to accurately estimate energy consumption and derive efficient attacks via Correlation Power Analysis (CPA). Furthermore, we provide an exploratory analysis of hyperparameter and weight leakage from LLMs in far field and demonstrate that the GPU's electromagnetic radiation leaks even 100 cm away through a glass obstacle.

cs.CR↗

arXiv2024

CNN architecture extraction on edge GPU

Neural networks have become popular due to their versatility and state-of-the-art results in many applications, such as image classification, natural language processing, speech recognition, forecasting, etc. These applications are also used in resource-constrained environments such as embedded devices. In this work, the susceptibility of neural network implementations to reverse engineering is explored on the NVIDIA Jetson Nano microcomputer via side-channel analysis. To this end, an architecture extraction attack is presented. In the attack, 15 popular convolutional neural network architectures (EfficientNets, MobileNets, NasNet, etc.) are implemented on the GPU of Jetson Nano and the electromagnetic radiation of the GPU is analyzed during the inference operation of the neural networks. The results of the analysis show that neural network architectures are easily distinguishable using deep learning-based side-channel analysis.

cs.CR↗

arXiv2023

BarraCUDA: Edge GPUs do Leak DNN Weights

Over the last decade, applications of neural networks (NNs) have spread to various aspects of our lives. A large number of companies base their businesses on building products that use neural networks for tasks such as face recognition, machine translation, and self-driving cars. Much of the intellectual property underpinning these products is encoded in the exact parameters of the neural networks. Consequently, protecting these is of utmost priority to businesses. At the same time, many of these products need to operate under a strong threat model, in which the adversary has unfettered physical control of the product. In this work, we present BarraCUDA, a novel attack on general purpose Graphic Processing Units (GPUs) that can extract parameters of neural networks running on the popular Nvidia Jetson Nano device. BarraCUDA uses correlation electromagnetic analysis to recover parameters of real-world convolutional neural networks.

cs.CR↗

arXiv2021

Surface Segmentation Using Implicit Divergence Constraint Between Adjacent Minimal Paths

We introduce a novel approach for object segmentation from 3D images using modified minimal path Eikonal equation. The proposed method utilizes an implicit constraint - a second order correction to the inhomogeneous minimal path Eikonal - preventing the adjacent minimal path trajectories to diverge uncontrollably. The proposed modification greatly reduces the surface area uncovered by minimal paths allowing the use of the calculated minimal path set as parameter lines of an approximate surface. It also has a loose connection with the true minimal surface Eikonal equations that are also deduced.

cs.CV↗

arXiv2019

An Elastic Energy Minimization Framework for Mean Surface Calculation

As the continuation of the contour mean calculation - designed for averaging the manual delineations of 3D layer stack images - in this paper, the most important equations: a) the reparameterization equations to determine the minimizing diffeomorphism and b) the proper centroid calculation for the surface mean calculation are presented. The chosen representation space: escaled Position by Square root Normal (RPSN) is a real valued vector space, invariant under the action of the reparameterization group and the imposed L2 metric (used to define the distance function) has well defined meaning: the sum of the central second moments of the coordinate functions. For comparision purpose, the reparameterization equations for elastic surface matching, using the Square Root Normal Function (SRNF) are also provided. The reparameterization equations for these cases have formal similarity, albeit the targeted applications differ: SRNF representation suitable for shape analysis purpose whereas RPSN is more fit for the cases where all contextual information - including the relative translation between the constituent surfaces - are to be retained (but the sake of theoretical completeness, the possibility of the consistent relative displacement removal in the RPSN case is also addressed).

math.MG↗

arXiv2019

An Elastic Energy Minimization Framework for Mean Contour Calculation

In this paper we propose a contour mean calculation and interpolation method designed for averaging manual delineations of objects performed by experts and interpolate 3D layer stack images. The proposed method retains all visible information of the input contour set: the relative positions, orientations and size, but allows invisible quantities - parameterization and the centroid - to be changed. The chosen representation space - the position vector rescaled by square root velocity - is a real valued vector space on which the imposed L2 metric is used to define the distance function. With respect to this representation the re-parameterization group acts by isometries and the distance has well defined meaning: the sum of the central second moments of the coordinate functions. To identify the optimal re-parameterization system and proper centroid we use double energy minimization realized in a variational framework.

math.MG↗

arXiv2018

modAL: A modular active learning framework for Python

modAL is a modular active learning framework for Python, aimed to make active learning research and practice simpler. Its distinguishing features are (i) clear and modular object oriented design (ii) full compatibility with scikit-learn models and workflows. These features make fast prototyping and easy extensibility possible, aiding the development of real-life active learning pipelines and novel algorithms as well. modAL is fully open source, hosted on GitHub at https://github.com/cosmic-cortex/modAL. To assure code quality, extensive unit tests are provided and continuous integration is applied. In addition, a detailed documentation with several tutorials are also available for ease of use. The framework is available in PyPI and distributed under the MIT license.

cs.LG↗

arXiv2018

Balanced Phase Field model for Active Surfaces

In this paper we present a balanced phase field model for active surfaces. This work is devoted to the generalization of the Balanced Phase Field Model for Active Contours devised to eliminate the often undesirable curvature-dependent shrinking of the zero level set while maintaining the smooth interface necessary to calculate the fundamental geometric quantities of the represented contour. As its antecedent work, the proposed model extends the Ginzburg-Landau phase field energy with a higher order smoothness term. The relative weights are determined with the analysis of the level set motion in a curvilinear system adapted to the zero level set. The proposed model exhibits strong shape maintaining capability without significant interference with the active (e.g. a segmentation) model.

math.OC↗

arXiv2006

A higher-order active contour model of a `gas of circles' and its application to tree crown extraction

Many image processing problems involve identifying the region in the image domain occupied by a given entity in the scene. Automatic solution of these problems requires models that incorporate significant prior knowledge about the shape of the region. Many methods for including such knowledge run into difficulties when the topology of the region is unknown a priori, for example when the entity is composed of an unknown number of similar objects. Higher-order active contours (HOACs) represent one method for the modelling of non-trivial prior knowledge about shape without necessarily constraining region topology, via the inclusion of non-local interactions between region boundary points in the energy defining the model. The case of an unknown number of circular objects arises in a number of domains, e.g. medical, biological, nanotechnological, and remote sensing imagery. Regions composed of an a priori unknown number of circles may be referred to as a `gas of circles'. In this report, we present a HOAC model of a `gas of circles'. In order to guarantee stable circles, we conduct a stability analysis via a functional Taylor expansion of the HOAC energy around a circular shape. This analysis fixes one of the model parameters in terms of the others and constrains the rest. In conjunction with a suitable likelihood energy, we apply the model to the extraction of tree crowns from aerial imagery, and show that the new model outperforms other techniques.

cs.CV↗