Conference Papers

You can also find my articles on my Google Scholar profile.

EvoGrad: Accelerated Metaheuristics in a Differentiable Wonderland

Published in IEEE Congress on Evolutionary Computation (CEC), 2026

EvoGrad is a unified differentiable framework that turns evolutionary and swarm operators into differentiable operators, integrating Evolutionary Computation and Swarm Intelligence with gradient-based optimization via backpropagation.

Recommended citation: Citterio B.F.R., Papetti D.M., Dimitri G.M., Tangherloni A. (2026). EvoGrad: Accelerated Metaheuristics in a Differentiable Wonderland. In IEEE Congress on Evolutionary Computation (CEC), IEEE.
Download Paper

Now You See Me, Now You Don’t: A Unified Framework for Expression Consistent Anonymization in Talking Head Videos

Published in IEEE/CVF International Conference on Computer Vision (ICCV), 2025

To enable privacy-preserving video analysis, we propose AnonNET, a diffusion-based system that de-identifies faces yet maintains age, gender, pose, expression, and realistic temporal dynamics.

Recommended citation: Egin A., Tangherloni A., Dantcheva A. (2025). Now You See Me, Now You Don’t: A Unified Framework for Expression Consistent Anonymization in Talking Head Videos. In IEEE/CVF International Conference on Computer Vision (ICCV), IEEE.
Download Paper

Knowledge-Enriched Cell-Type Annotation in Single-Cell Transcriptomics via LLM Embeddings

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2025

We show that embedding biological knowledge through Modern-BERT gene vectors strengthens supervised models for scRNA-seq cell-type classification.

Recommended citation: Fabbricatore A., Buffa F.M., Tangherloni A. (2025). Knowledge-Enriched Cell-Type Annotation in Single-Cell Transcriptomics via LLM Embeddings. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

GENESIS: Generating scRNA-Seq data from Multiome Gene Expression

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2025

We introduce GENESIS, a generative modelling framework that bridges the gap between Multiome GEX and scRNA-Seq data by producing enhanced whole-cell–like expression profiles.

Recommended citation: Riva S.G., Myers B., Buffa F.M., Tangherloni A. (2025). GENESIS: Generating scRNA-Seq data from Multiome Gene Expression. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

We Are Sending You Back… to the Optimum! Fuzzy Time Travel Particle Swarm Optimization

Published in International Conference on the Applications of Evolutionary Computation (Part of EvoStar), 2025

To prevent FST-PSO from getting trapped in local optima, we propose FTT-PSO, a time-travel strategy that rewinds the swarm and reinitialises the global best when progress stalls, yielding superior results on major benchmark suites.

Recommended citation: Papetti D.M., Tangherloni A., Coelho V., Besozzi D., Cazzaniga P., Nobile M.S. (2025). We Are Sending You Back... to the Optimum! Fuzzy Time Travel Particle Swarm Optimization. In International Conference on the Applications of Evolutionary Computation (Part of EvoStar), Springer.
Download Paper

Forest-based Evolutionary Algorithm for Reconstructing Boolean Gene Regulatory Networks

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2024

We introduce FP, a novel forest-structured evolutionary algorithm that models complex gene regulatory interactions and reconstructs Boolean GRNs with efficient convergence and strong precision.

Recommended citation: Stranieri N., Buffa F.M.,Tangherloni A. (2024). Forest-based Evolutionary Algorithm for Reconstructing Boolean Gene Regulatory Networks. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

A Fast Feature Selection for Interpretable Modeling Based on Fuzzy Inference Systems

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2024

We propose a novel and efficient feature-selection strategy for Fuzzy Inference Systems that uses Random Forest–based variable ranking to identify the smallest set of informative features.

Recommended citation: Tangherloni A., Cazzaniga P., Stranieri N., Buffa F.M., Nobile M.S. (2024). A Fast Feature Selection for Interpretable Modeling Based on Fuzzy Inference Systems. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

A Modified EACOP Implementation for Real-Parameter Single Objective Optimization Problems

Published in IEEE Congress on Evolutionary Computation (CEC), 2024

We introduce iEACOP, an enhanced evolutionary algorithm with minimal hyper-parameter tuning requirements, which outperforms the original EACOP on most CEC 2017 benchmark functions and achieves performance comparable to top competition algorithms.

Recommended citation: Tangherloni A., Coelho V., Buffa F.M., Cazzaniga P. (2024). A Modified EACOP Implementation for Real-Parameter Single Objective Optimization Problems. In IEEE Congress on Evolutionary Computation (CEC), IEEE.
Download Paper

Consensus Clustering Strategy for Cell Type Assignments of scRNA-seq Data

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2023

We introduce a pseudo-voting consensus approach within scALPO that combines multiple clustering outputs to achieve more robust and accurate cell-type annotation.

Recommended citation: Riva S.G., Myers B., Cazzaniga P., Buffa F.M., Tangherloni A. (2023). Consensus Clustering Strategy for Cell Type Assignments of scRNA-seq Data. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

The Domination Game: Dilating Bubbles to Fill Up Pareto Fronts

Published in IEEE Congress on Evolutionary Computation (CEC), 2023

By manipulating the search space via LBDFs, our method reliably discovers new non-dominated solutions and improves Pareto front quality across standard multi-objective benchmarks.

Recommended citation: Coelho V., Papetti D.M., Tangherloni A., Cazzaniga P., Besozzi D., Nobile M.S. (2023). The Domination Game: Dilating Bubbles to Fill Up Pareto Fronts. In IEEE Congress on Evolutionary Computation (CEC), IEEE.
Download Paper

Multi-objective Optimization for Marker Panel Identification in Single-cell Data

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2022

To address the complexity of selecting minimal yet discriminative marker panels, we model the task as a bi-objective optimization problem and evaluate multi-objective algorithms that outperform genetic algorithms in both quality and consistency.

Recommended citation: Tangherloni A., Riva S.G., Myers B., Cazzaniga P. (2022). Multi-objective Optimization for Marker Panel Identification in Single-cell Data. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

A Deep Learning Pipeline for the Automatic cell type Assignment of scRNA-seq Data

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2022

scALPO is a novel automated pipeline that employs an LSTM neural network to assign cell types based solely on marker genes, surpassing current state-of-the-art tools in annotation accuracy.

Recommended citation: Riva S.G., Myers B., Cazzaniga P., Tangherloni A.(2022).A Deep Learning Pipeline for the Automatic cell type Assignment of scRNA-seq Data. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

Integration of Multiple scRNA-Seq Datasets on the Autoencoder Latent Space

Published in IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2021

A novel Autoencoder-based strategy for integrating multiple scRNA-Seq datasets that outperforms Scanorama, Ingest, and Seurat in most cases.

Recommended citation: Riva S.G., Cazzaniga P., Tangherloni A. (2021). Integration of Multiple scRNA-Seq Datasets on the Autoencoder Latent Space. In IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pp. 2155-2162, IEEE.
Download Paper

The impact of representation on the optimization of marker panels for single-cell RNA data

Published in IEEE Congress on Evolutionary Computation (CEC), 2021

To address the NP-hard challenge of marker panel identification, we introduce and compare three GA representations, demonstrating that more flexible encodings yield the most effective panels, especially in 0-knowledge scenarios.

Recommended citation: Tangherloni A., Riva S.G., Spolaor S., Besozzi D., Nobile M.S., Cazzaniga P. (2021). The impact of representation on the optimization of marker panels for single-cell RNA data. In IEEE Congress on Evolutionary Computation (CEC), IEEE.
Download Paper

Efficient and Settings-Free Calibration of Detailed Kinetic Metabolic Models with Enzyme Isoforms Characterization

Published in International Conference on Computational Intelligence Methods for Bioinformatics and Biostatistics (CIBB), 2020

An efficient, settings-free strategy for calibrating detailed kinetic metabolic models, including the characterization of enzyme isoforms.

Recommended citation: Totis N., Tangherloni A., Beccuti M., Cazzaniga P., Nobile M.S., Besozzi D., Pennisi M., Pappalardo F. (2020). Efficient and Settings-Free Calibration of Detailed Kinetic Metabolic Models with Enzyme Isoforms Characterization. In Computational Intelligence Methods for Bioinformatics and Biostatistics (CIBB), Lecture Notes in Computer Science, 11925: 187-202, Springer.
Download Paper

HaraliCU: GPU-powered Haralick Feature Extraction on Medical Images Exploiting the Full Dynamics of Gray-Scale Levels

Published in International Conference on Parallel Computing Technologies (PaCT), 2019

HaraliCU is a GPU-powered method for Haralick feature extraction on medical images that exploits the full dynamics of gray-scale levels without requantization.

Recommended citation: Rundo L., Tangherloni A., Galimberti S., Cazzaniga P., Woitek R., Sala E., Nobile M.S., Mauri G. (2019). HaraliCU: GPU-powered Haralick Feature Extraction on Medical Images Exploiting the Full Dynamics of Gray-Scale Levels. In International Conference on Parallel Computing Technologies (PaCT), Lecture Notes in Computer Science, 11657: 304-318, Springer.
Download Paper

GPU accelerated analysis of Treg-Teff cross regulation in relapsing-remitting multiple sclerosis

Published in Euro-Par 2018: Parallel Processing Workshops (Springer), 2019

GPU-accelerated analysis of the cross regulation between regulatory and effector T cells in relapsing-remitting multiple sclerosis.

Recommended citation: Beccuti M., Cazzaniga P., Pennisi M., Besozzi D., Nobile M.S., Pernice S., Russo G., Tangherloni A., Pappalardo F. (2019). GPU accelerated analysis of Treg-Teff cross regulation in relapsing-remitting multiple sclerosis. In Euro-Par 2018: Parallel Processing Workshops, Lecture Notes in Computer Science, 11339: 626-637, Springer.
Download Paper

High Performance Computing for Haplotyping: Models and Platforms

Published in Euro-Par 2018: Parallel Processing Workshops (Springer), 2019

A review of models and high-performance computing platforms for the haplotype assembly problem.

Recommended citation: Tangherloni A., Rundo L., Spolaor S., Nobile M.S., Merelli I., Besozzi D., Mauri G., Cazzaniga P., Liò P. (2019). High Performance Computing for Haplotyping: Models and Platforms. In Euro-Par 2018: Parallel Processing Workshops, Lecture Notes in Computer Science, 11339: 650-661, Springer.
Download Paper

Estimation of Kinetic Reaction Constants: Exploiting Reboot Strategies to Improve PSO’s Performance

Published in International Conference on Computational Intelligence Methods for Bioinformatics and Biostatistics (CIBB), 2019

Reboot strategies for Particle Swarm Optimization that help avoid local optima when estimating kinetic reaction constants of biochemical models.

Recommended citation: Spolaor S., Tangherloni A., Rundo L., Cazzaniga P., Nobile M.S. (2019). Estimation of Kinetic Reaction Constants: Exploiting Reboot Strategies to Improve PSO’s Performance. In Computational Intelligence Methods for Bioinformatics and Biostatistics (CIBB), Lecture Notes in Computer Science, 10834: 92-102, Springer.
Download Paper

Computational Intelligence for Parameter Estimation of Biochemical Systems

Published in IEEE Congress on Evolutionary Computation (CEC), 2018

To address the challenge of estimating kinetic parameters in biochemical models, we benchmark seven state-of-the-art optimization techniques and find that a settings-free FST-PSO variant yields the most robust and accurate performance across diverse systems.

Recommended citation: Nobile M.S., Tangherloni A., Rundo L., Spolaor S., Besozzi D., Mauri G., Cazzaniga P. (2018). Computational Intelligence for Parameter Estimation of Biochemical Systems. In IEEE Congress on Evolutionary Computation (CEC), IEEE.
Download Paper

GPU-Powered Multi-Swarm Parameter Estimation of Biological Systems: A Master-Slave Approach

Published in Euromicro International Conference on Parallel, Distributed, and Network-based Processing (PDP), 2018

MS2PSO is a parallel and distributed multi-swarm PSO, powered by GPU simulation, for estimating reaction constants in biological models across multiple experimental conditions.

Recommended citation: Tangherloni A., Rundo L., Spolaor S., Cazzaniga P., Nobile M.S. (2018). GPU-Powered Multi-Swarm Parameter Estimation of Biological Systems: A Master-Slave Approach. In IEEE 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP), pp. 698-705, IEEE.
Download Paper

Reboot Strategies in Particle Swarm Optimization and their Impact on Parameter Estimation of Biochemical Systems

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2017

Three reboot strategies for PSO that reinitialize particle positions to avoid local optima in the parameter estimation of biochemical systems, accelerated with GPU simulation.

Recommended citation: Spolaor S., Tangherloni A., Rundo L., Nobile M.S., Cazzaniga P. (2017). Reboot Strategies in Particle Swarm Optimization and their Impact on Parameter Estimation of Biochemical Systems. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

Proactive Particles in Swarm Optimization: A settings-free algorithm for real-parameter single objective optimization problems

Published in IEEE Congress on Evolutionary Computation (CEC), 2017

To address PSO’s sensitivity to parameter settings, we propose PPSO, a fuzzy logic–driven algorithm that proactively adapts parameters for each particle and improves convergence and performance across high-dimensional benchmarks.

Recommended citation: Tangherloni A., Rundo L., Nobile M.S. (2017). Proactive Particles in Swarm Optimization: A settings-free algorithm for real-parameter single objective optimization problems. In IEEE Congress on Evolutionary Computation (CEC), IEEE.
Download Paper

Multimodal medical image registration using Particle Swarm Optimization: A review

Published in IEEE Symposium Series on Computational Intelligence (SSCI), 2016

A critical review of biomedical image registration using Particle Swarm Optimization and its hybridizations with Evolutionary Strategies.

Recommended citation: Rundo L., Tangherloni A., Militello C., Gilardi M.C., Mauri G. (2016). Multimodal medical image registration using Particle Swarm Optimization: A review. In IEEE Symposium Series on Computational Intelligence (SSCI), IEEE.
Download Paper

GPU-powered bat algorithm for the parameter estimation of biochemical kinetic values

Published in IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2016

A Lévy-flight Bat Algorithm combined with GPU simulation outperforms PSO in accuracy and speed for biochemical parameter estimation.

Recommended citation: Tangherloni A., Nobile M.S., Cazzaniga P. (2016). GPU-powered bat algorithm for the parameter estimation of biochemical kinetic values. In IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE.
Download Paper

GPU-powered and settings-free parameter estimation of biochemical systems

Published in IEEE Congress on Evolutionary Computation (CEC), 2016

TTo address the computational demands of biochemical parameter estimation, we combine PPSO with GPU-accelerated ODE simulation and demonstrate that PPSO outperforms standard PSO in runtime while achieving similar parameter-fitness quality.

Recommended citation: Nobile M.S., Tangherloni A. Besozzi D., Cazzaniga P. (2016). GPU-powered and settings-free parameter estimation of biochemical systems. In IEEE Congress on Evolutionary Computation (CEC), IEEE.
Download Paper

Andrea Tangherloni

Conference Papers