L. Chicchi, D. Fanelli, D. Febbe, L. Buffoni, F. Di Patti, L. Giambagli, Raffaele Marino. Deterministic versus stochastic dynamical classifiers: opposing random adversarial attacks with noise. Machine Learning: Science and Technology, (2025). Article link.
L. Serricchio, C. Chilin, D. Bocchi, Raffaele Marino, M. Negri, C. Cammarota, F. Ricci-Tersenghi. Daydreaming Hopfield Networks and their surprising effectiveness on correlated data. Neural Networks, 107216, (2025). Article link.
Raffaele Marino, L. Buffoni, L. Chicchi, F. Di Patti, D. Febbe, L. Giambagli, D. Fanelli. Learning in the Wilson-Cowan model for metapopulation. Neural Computation 37:4, (2025). Article link.
Raffaele Marino, L. Giambagli, L. Chicchi, L. Buffoni, D. Fanelli. Stable Attractors for Neural networks classification via Ordinary Differential Equations (SA-nODE). Machine Learning: Science and Technology, 5, 035087, (2024). Article link.
Maria Chiara Angelini, Angelo Giorgio Cavaliere, Raffaele Marino, F. Ricci-Tersenghi. Stochastic Gradient Descent-like relaxation is equivalent to Metropolis dynamics in discrete optimization and inference problems. Sci. Rep. 14, 11638 (2024). Article link.
L. Chicchi, L. Giambagli, L. Buffoni, Raffaele Marino, D. Fanelli. Complex Recurrent Spectral Network. Chaos, Solitons & Fractals. 184, 114998 (2024). Article link.
Raffaele Marino, F. Ricci-Tersenghi. Phase transitions in the mini-batch size for sparse and dense two-layer neural networks. Machine Learning: Science and Technology 5, 015015 (2024). Article link.
C. Nicolini, J. Staiano, B. Lepri, Raffaele Marino. The Garden of Forking Paths: Observing Dynamic Parameters Distribution in Large Language Models. arXiv:2403.08739 (2024). Article link.
L. Chicchi, L. Buffoni, D. Febbe, L. Giambagli, Raffaele Marino, D. Fanelli. Automatic Input Feature Relevance via Spectral Neural Networks. arXiv:2406.01183 (2024). Article link.
Raffaele Marino. Fast Analysis of the OpenAI O1-Preview Model in Solving Random K-SAT Problem: Does the LLM Solve the Problem Itself or Call an External SAT Solver?. arXiv:2409.11232 (2024). Article link.
Raffaele Marino. A Bridge between Dynamical Systems and Machine Learning: Engineered Ordinary Differential Equations as Classification Algorithm (EODECA). BULLETIN OF THE AMERICAN PHYSICAL SOCIETY, ISSN: 0003-0503 (2024). Article link.
Raffaele Marino, N. Macris. Solving non-linear Kolmogorov equations by using deep learning: a numerical comparison of discretization schemes. Journal of Scientific Computing 94,8 (2023):1-31. Article link.
Ludovica Serricchio, Claudio Chilin, Dario Bocchi, Raffaele Marino, Matteo Negri, Chiara Cammarota, Federico Ricci-Tersenghi. Daydreaming Hopfield Networks and their surprising effectiveness on correlated data. Associative Memory & Hopfield Networks, NeurIPS 2023. p. 1-8, OpenReview.net, New Orleans USA, 11-16/12. Article link.
Raffaele Marino. Learning from Survey Propagation: a Neural Network for MAX-E-3-SAT. Machine Learning: Science and Technology 2.3 (2021): 035032. Article link.
Beatrice Donelli, Stefano Gherardini, Raffaele Marino, Francesco Campaioli, Lorenzo Buffoni. Charging a quantum spin network with superextensive precision. Physical Review E, 111, L062102 (2025). Article link.
Marco Baldovin, Raffaele Marino, Angelo Vulpiani. Ergodic observables in non-ergodic systems: the example of the harmonic chain. Phys. A: Stat. Mech. its Appl. 129273 (2023) . Article link.
S. Caracciolo, R. Fabbricatore, Raffaele Marino, G. Parisi, G. Sicuro. Criticality and conformality in the random dimer problem. Physical Review E 103.4 (2021): 042127. Article link.
Raffaele Marino, R. Eichhorn, Brownian motion of an ellipsoidal particle in a tilted periodic potential: long term velocity and diffusion. (2017) . Article link.
Raffaele Marino, Dynamics and Thermodynamics of Translational and Rotational Diffusion Processes Driven out of equilibrium. KTH, School of Computer Science and Communications. ISBN: 978-91-7729-405-4, 2017. Article link.
Raffaele Marino, E. Aurell. Advective-diffusive motion on large scale from small scale dynamics with an internal symmetry. Physical Review E 93.6 (2016): 062147. Article link.
E. Aurell, S. Bo, M. Dias, R. Eichhorn, Raffaele Marino. Diffusion of Brownian ellipsoid in a force field. EPL (Europhysics Letters) 114.3 (2016): 30005. Article link.
Raffaele Marino, R. Eichhorn, E. Aurell. Entropy production of a Brownian ellipsoid in the overdamped limit. Physical Review E 93.1 (2016): 012132. Article link.
Raffaele Marino, L. Buffoni, B. Zavalnij. A Short Review on Novel Approaches for Maximum Clique Problem: from Classical algorithms to Graph Neural Networks and Quantum algorithms arXiv:2403.09742 (2024). Article link.
Raffaele Marino, S. Kirkpatrick. Hard Optimization Problems have Soft Edges. Sci. Rep. 13, 3671 (2023). Article link.
Raffaele Marino, S. Kirkpatrick. Large independent set on random d-regular graphs with fixed degree d. Computation 11, 206 (2023). Article link.
Raffaele Marino. Where do hard problems really exists? arXiv:2309.16253 (2023). Article link.
Masoud Mohseni, Daniel Eppens, Federico Ricci-Tersenghi, Johan Strumpfer, Alan Ho, Raffaele Marino, Vasil Denchev, Sergei Isakov, Sergio Boixo, Hartmut Neven (2022). Nonequilibrium Monte Carlo for unfreezing variables near computational phase transitions. BULLETIN OF THE AMERICAN PHYSICAL SOCIETY, ISSN: 0003-0503. Article link.
M. Mohseni, D. Eppens, J. Strumpfer, Raffaele Marino, V. Denchev, A. K. Ho, S. V. Isakov, S. Boixo, F. Ricci-Tersenghi, H. Neven Nonequilibrium Monte Carlo for unfreezing variables in hard combinatorial optimization. arXiv:2111.13628 (2021). Article link.
Raffaele Marino, S. Kirkpatrick. Revisiting the Challenges of MaxClique. arXiv:1807.09091 (2018) . Article link.
Raffaele Marino, G. Parisi, F. Ricci-Tersenghi. The backtracking survey propagation algorithm for solving K-SAT problems. Nature communications 7.1 (2016): 1-8. Article link.
Invited talk: Neural Mass Network Models as Classifiers. 19/03/2025 GGI, Firenze, Italy.
Invited talk: Learning in Wilson and Cowan Model for metapopulation. BIOPHYS2024, 11/09/2024, Firenze, Italy
Invited talk: "AI \& Deep Learning: from Alan Turing to Chat-GPT". GARR conference 2023. 29/05/2023, Firenze, Italy.
Invited talk:Hard Optimization Problems have Soft Edges. SummerSolstice conference 2023. 12/04/2023, Firenze, Italy.
Invited talk: Message passing algorithms \& greedy algorithms: two different approaches for solving constraint satisfaction problems. PostDoc Mini-Symposium. 13/12/2018 The Hebrew University of Jerusalem, Faculty of Science, Jerusalem, Israel.
Invited talk: Entropy production of a Brownian ellipsoid in the overdamped limit. NORDITA Day-Winter 2015. 20/11/2015 NORDITA, Stockholm, Sweden.
Contributed talk: Neural Mass Network Models as Classifiers. 13-18/07/2025 StatPhys29, Firenze, Italy.
Contributed talk: Wilson and Cowan Model as a Learning Algorithm: a step towards biologically inspired machine learning algorithms. MNESYS Spoke 2, Neural Plasticity and Connectivity, 3/05/2024, Napoli, Italy
Contributed talk: A Bridge between Dynamical Systems and Machine Learning: Engineered Ordinary Differential Equations as Classification Algorithm (EODECA). APS March Meetings 3-8/03/2024, Minneapoilis, USA
Contributed talk: Phase transitions in the mini-batch size for sparse and dense deep neural networks. International Conference in Statistical Physics. 10-14/07/2023, Chania-Crete, Greece
Contributed talk: Solving non-linear Kolmogorov equations by using deep learning: a numerical comparison of discretization schemes. Santa Fe 3rd Physics Informed Machine Learning. 13/01/2020, Santa Fe, USA (NM)
Contributed talk: A greedy story of Max-Clique. Mathematical and computational aspects of machine learning. 9/10/2019, Scuola Normale di Pisa, Pisa, Italy.
Poster: Wilson and Cowan Model as a Learning Algorithm: a step towards biologically inspired machine learning algorithms. MNESYS Spoke 2, Neural Plasticity and Connectivity, 3/05/2024, Napoli, Italy
Poster: Engineered Ordinary Differential Equations as Classification Algorithm (EODECA): a Bridge between Dynamical Systems and Machine Learning. Complex systems, statistical mechanics and machine learning crossover. 24-29/03/2024 Les Houches, France.
Poster: Engineered Ordinary Differential Equations as Classification Algorithm (EODECA): a Bridge between Dynamical Systems and Machine Learning. NetSci-X 2024 Conference. 22-25/01/2024 Venice, Italy.
Poster: Revisiting the challenges of MaxClique. Statistical physics and machine learning back together. 24/8/2018 Cargese, France.
Poster: Diffusion of a Brownian ellipsoid in a force field. Statistical mechanics of quantum dynamics. 5/2016 Mariehamn, Sweden.
Invited talk: Neural Mass Network Models as Classifiers. 10/12/2024 Human Technopole, Milano, Italy.
Invited talk: Engineered Ordinary Differential Equations as Classification Algorithm (EODECA): a Bridge between Dynamical Systems and Machine Learning. 14/03/2024 Università degli studi di Firenze, Dipartimento di Fisica e Astronomia, Firenze, Italy.
Invited talk: Hard Optimization Problems have Soft Edges. 14/06/2023 Università degli studi di Firenze, Dipartimento di Fisica e Astronomia, Firenze, Italy.
Invited talk: Deep Learning: a personal introduction. 05/11/2021 Università degli studi di Roma La Sapienza, Dipartimento di Fisica, Roma, Italy.
Invited talk: Solving non-linear Kolmogorov equations by using deep learning: a numerical comparison of discretization schemes. Complex systems and Biological physics seminar. 08/10/2020 NORDITA, Stockholm, Sweden.
Invited talk: Solving non-linear Kolmogorov equations by using deep learning: a numerical comparison of discretization schemes. Los Alamos National Laboratories Seminar. 23/01/2020 Los Alamos National Laboratories, USA (NM).
Invited talk: Revisiting the challenges of MaxClique. IPG Seminar. 17/10/2018 EPFL, School of Computer and Comm. Science, Lausanne, Switzerland.
Invited talk: Dynamics and thermodynamics of translational and rotational diffusion processes driven out of equilibrium. Ph.D. Thesis public defence. 15/6/2017 Royal Institute of Technology (KTH), Stockholm, Sweden.
Invited talk: The Backtracking Survey Propagation algorithm for solving random K-SAT problems. Complex systems and Biological physics seminar. 20/9/2016 NORDITA, Stockholm, Sweden.
Invited talk: Diffusion of a Brownian ellipsoid in a force field. Complex systems and Biological physics seminar. 23/8/2016 NORDITA, Stockholm, Sweden.
Invited talk: Anomalous entropy production of a Brownian particle: the general case. Complex systems and Biological physics seminar. 09/12/2014 NORDITA, Stockholm, Sweden.
Invited talk: Optimization algorithms for K-SAT problems. Complex systems and Biological physics seminar. 23/10/2013 NORDITA, Stockholm, Sweden.