Institution profile

Jagiellonian University

Academic institutioneurope · pl
Official website
Research library232linked papers
Opportunities0open roles
Selected work

Representative Papers

KNOWM memristors in a bridge synapse delay-based reservoir computing system for detection of epileptic seizures

Jun 26, 2022Int. J. Parallel Emergent Distributed Syst.

To address the need for real-time, low-power epileptic seizure detection, this work proposes a novel memristor-based bridge synaptic delayed reservoir computing architecture using KNOWM memristors. The method integrates four KNOWM memristors with a differential amplifier to construct a single-node echo state machine (SNESM) as a physical reservoir, leveraging feedback loops to perform nonlinear temporal transformation and disentangle complexity features from raw EEG signals. This represents the first hardware implementation of a memristive bridge synaptic structure for delayed reservoir computing, significantly enhancing discriminability among complexity metrics across distinct epileptic states. Experimental results demonstrate reduced inter-feature correlation and improved class separability in the transformed signal space compared to raw EEG, leading to markedly higher seizure detection accuracy. The approach establishes a new paradigm for neuromorphic edge-intelligent healthcare systems.

6 citations1 influentialRead paper

Scalable and Cost-Efficient de Novo Template-Based Molecular Generation

Jun 10, 2025arXiv.org

This work addresses three key challenges in template-guided molecular generation: high synthetic cost, difficulty in scaling the building block library, and underutilization of small fragments. We propose a recursive, cost-guided generative framework based on Generative Flow Networks (GFlowNets). Methodologically, we design a backward policy network coupled with an auxiliary synthetic cost predictor, introduce a dynamic building block library that reuses intermediate molecular states, and employ a penalty mechanism to balance exploration and exploitation. Our key contribution lies in explicitly embedding synthetic cost into the generative process, enabling end-to-end differentiable optimization; the dynamic library mechanism markedly improves both diversity and efficiency—especially with limited building block sets. On standard templated molecular generation benchmarks, our approach generates higher-quality, more diverse molecules at lower synthetic cost, achieving state-of-the-art performance.

2 citationsRead paper

Basis Number of Graphs Excluding Minors

Jan 08, 2026

This study investigates whether the basis number of graphs excluding a fixed minor $H$ is bounded—that is, whether their cycle space admits a generating set of cycles in which each edge appears at most a constant number of times. By integrating graph minor exclusion theory with tree and path decomposition techniques based on the Bojańczyk–Pilipczuk framework, the work establishes the first connection between basis number and minor exclusion. The main contribution is a proof that every $H$-minor-free graph has a bounded basis number, together with an explicit polynomial upper bound of the form $O(|H|^c)$, where $c$ is an absolute constant. This result significantly extends beyond prior bounds that were restricted to graphs embeddable on surfaces.

1 citationsRead paper

Nonuniform Deterministic Finite Automata over finite algebraic structures

Jan 21, 2025

This paper investigates the modeling power and computational complexity of non-uniform deterministic finite automata (NUDFA) over finite algebraic structures, focusing on three fundamental problems: identity checking, equation satisfiability, and circuit equivalence testing. We first generalize the semantics of NUDFA to arbitrary finite algebras, establishing a formal semantic framework and a unified decision-theoretic foundation. Second, we fully characterize the class of finite groups admitting probabilistic polynomial-time identity checking—namely, those whose Sylow subgroups are all abelian. Third, we identify an algebraic characterization for efficient circuit equivalence testing within modular equivalence varieties: the problem is in BPP precisely when the underlying algebra belongs to a variety that is both idempotent and abelian. Our results integrate algebraic automata theory, group theory, randomized algorithms, and complexity-theoretic assumptions (rETH/CDH), yielding a novel paradigm and tight complexity-theoretic characterization for automata-based computation over finite algebras.

1 citationsRead paper

MiraGe: Editable 2D Images using Gaussian Splatting

Oct 02, 2024arXiv.org

This work addresses the challenge of achieving high-quality, physically consistent, and semantically editable 2D image editing within 3D space. We propose a semantic-controllable image editing framework built upon 3D Gaussian representations. Methodologically, we introduce the first integration of specular reflection geometry modeling with planar-constrained Gaussian splatting to construct a differentiable and editable implicit 3D scene representation; this is further coupled with a physics engine to ensure illumination consistency, geometrically plausible occlusion, and physically grounded dynamics during interactive editing. Compared to conventional implicit neural representations (INRs) and non-editable Gaussian rendering approaches, our method overcomes the semantic-level manipulation bottleneck. It achieves state-of-the-art compression efficiency while significantly improving editing fidelity and 3D geometric consistency—establishing a novel paradigm for photorealistic image editing.

1 citationsRead paper
Recent publications

Latest Papers

Diversity of EML-type operators

Sep 10, 2026

本文通过列举和分类EML型操作符，澄清常见误解，并提出Möbius层及新的激活函数eml(x,1/x)，以解决在神经网络架构中实现基本函数评估的问题。

0 citationsRead paper