Skip to main content

Showing 1–29 of 29 results for author: Grudinin, S

.
  1. arXiv:2603.16739  [pdf, ps, other

    cs.LG cs.AI cs.HC

    SpecMoE: Spectral Mixture-of-Experts Foundation Model for Cross-Species EEG Decoding

    Authors: Davy Darankoum, Chloé Habermacher, Julien Volle, Sergei Grudinin

    Abstract: Decoding the orchestration of neural activity in electroencephalography (EEG) signals is a central challenge in bridging neuroscience with artificial intelligence. Foundation models have made strides in generalized EEG decoding, yet many existing frameworks primarily relying on separate temporal and spectral masking of raw signals during self-supervised pretraining. Such strategies often tend to b… ▽ More

    Submitted 30 March, 2026; v1 submitted 17 March, 2026; originally announced March 2026.

    Comments: 34 pages (12 pages in the main text and 22 pages in Supplementary Information)

  2. arXiv:2509.20489  [pdf, ps, other

    cs.LG cs.AI

    CoSupFormer : A Contrastive Supervised learning approach for EEG signal Classification

    Authors: D. Darankoum, C. Habermacher, J. Volle, S. Grudinin

    Abstract: Electroencephalography signals (EEGs) contain rich multi-scale information crucial for understanding brain states, with potential applications in diagnosing and advancing the drug development landscape. However, extracting meaningful features from raw EEG signals while handling noise and channel variability remains a major challenge. This work proposes a novel end-to-end deep-learning framework th… ▽ More

    Submitted 24 September, 2025; originally announced September 2025.

    Comments: 20 pages (14 pages Main text and 6 pages Supplementary Material)

  3. arXiv:2509.00832  [pdf, ps, other

    cs.LG cond-mat.mtrl-sci

    Challenges in Non-Polymeric Crystal Structure Prediction: Why a Geometric, Permutation-Invariant Loss is Needed

    Authors: Emmanuel Jehanno, Romain Menegaux, Julien Mairal, Sergei Grudinin

    Abstract: Crystalline structure prediction is an essential prerequisite for designing materials with targeted properties. Yet, it is still an open challenge in materials design and drug discovery. Despite recent advances in computational materials science, accurately predicting three-dimensional non-polymeric crystal structures remains elusive. In this work, we focus on the molecular assembly problem, where… ▽ More

    Submitted 26 September, 2025; v1 submitted 31 August, 2025; originally announced September 2025.

  4. arXiv:2504.02839  [pdf, other

    q-bio.BM cs.LG

    PETIMOT: A Novel Framework for Inferring Protein Motions from Sparse Data Using SE(3)-Equivariant Graph Neural Networks

    Authors: Valentin Lombard, Sergei Grudinin, Elodie Laine

    Abstract: Proteins move and deform to ensure their biological functions. Despite significant progress in protein structure prediction, approximating conformational ensembles at physiological conditions remains a fundamental open problem. This paper presents a novel perspective on the problem by directly targeting continuous compact representations of protein motions inferred from sparse experimental observa… ▽ More

    Submitted 19 March, 2025; originally announced April 2025.

    Journal ref: LMRL Workshop at ICLR 2025

  5. arXiv:2502.21274  [pdf, ps, other

    cs.LG cs.AI q-bio.BM

    BAnG: Bidirectional Anchored Generation for Conditional RNA Design

    Authors: Roman Klypa, Alberto Bietti, Sergei Grudinin

    Abstract: Designing RNA molecules that interact with specific proteins is a critical challenge in experimental and computational biology. Existing computational approaches require a substantial amount of previously known interacting RNA sequences for each specific protein or a detailed knowledge of RNA structure, restricting their utility in practice. To address this limitation, we develop RNA-BAnG, a deep… ▽ More

    Submitted 23 June, 2025; v1 submitted 28 February, 2025; originally announced February 2025.

  6. From Epilepsy Seizures Classification to Detection: A Deep Learning-based Approach for Raw EEG Signals

    Authors: Davy Darankoum, Manon Villalba, Clelia Allioux, Baptiste Caraballo, Carine Dumont, Eloise Gronlier, Corinne Roucard, Yann Roche, Chloe Habermacher, Sergei Grudinin, Julien Volle

    Abstract: Epilepsy represents the most prevalent neurological disease in the world. One-third of people suffering from mesial temporal lobe epilepsy (MTLE) exhibit drug resistance, urging the need to develop new treatments. A key part in anti-seizure medication (ASM) development is the capability of detecting and quantifying epileptic seizures occurring in electroencephalogram (EEG) signals, which is crucia… ▽ More

    Submitted 2 November, 2025; v1 submitted 4 October, 2024; originally announced October 2024.

    Comments: 25 pages, 3 tables, 5 figures

  7. arXiv:2404.15979  [pdf, other

    cs.CV cs.LG math.GR

    On the Fourier analysis in the SO(3) space : EquiLoPO Network

    Authors: Dmitrii Zhemchuzhnikov, Sergei Grudinin

    Abstract: Analyzing volumetric data with rotational invariance or equivariance is an active topic in current research. Existing deep-learning approaches utilize either group convolutional networks limited to discrete rotations or steerable convolutional networks with constrained filter structures. This work proposes a novel equivariant neural network architecture that achieves analytical Equivariance to Loc… ▽ More

    Submitted 19 March, 2025; v1 submitted 24 April, 2024; originally announced April 2024.

    Journal ref: conference paper at ICLR 2025

  8. ILPO-NET: Network for the invariant recognition of arbitrary volumetric patterns in 3D

    Authors: Dmitrii Zhemchuzhnikov, Sergei Grudinin

    Abstract: Effective recognition of spatial patterns and learning their hierarchy is crucial in modern spatial data analysis. Volumetric data applications seek techniques ensuring invariance not only to shifts but also to pattern rotations. While traditional methods can readily achieve translational invariance, rotational invariance possesses multiple challenges and remains an active area of research. Here,… ▽ More

    Submitted 24 April, 2024; v1 submitted 28 March, 2024; originally announced March 2024.

    Journal ref: Machine Learning and Knowledge Discovery in Databases. Research Track. ECML PKDD 2024. Lecture Notes in Computer Science(), vol 14944. Springer, Cham

  9. arXiv:2107.12078  [pdf, other

    q-bio.QM cs.LG eess.SP

    6DCNN with roto-translational convolution filters for volumetric data processing

    Authors: Dmitrii Zhemchuzhnikov, Ilia Igashov, Sergei Grudinin

    Abstract: In this work, we introduce 6D Convolutional Neural Network (6DCNN) designed to tackle the problem of detecting relative positions and orientations of local patterns when processing three-dimensional volumetric data. 6DCNN also includes SE(3)-equivariant message-passing and nonlinear activation operations constructed in the Fourier space. Working in the Fourier space allows significantly reducing t… ▽ More

    Submitted 30 July, 2021; v1 submitted 26 July, 2021; originally announced July 2021.

  10. arXiv:2105.07407  [pdf, other

    q-bio.BM cs.LG

    Protein sequence-to-structure learning: Is this the end(-to-end revolution)?

    Authors: Elodie Laine, Stephan Eismann, Arne Elofsson, Sergei Grudinin

    Abstract: The potential of deep learning has been recognized in the protein structure prediction community for some time, and became indisputable after CASP13. In CASP14, deep learning has boosted the field to unanticipated levels reaching near-experimental accuracy. This success comes from advances transferred from other machine learning areas, as well as methods specifically designed to deal with protein… ▽ More

    Submitted 13 September, 2021; v1 submitted 16 May, 2021; originally announced May 2021.

  11. arXiv:2011.07980  [pdf, other

    q-bio.QM cs.LG

    Spherical convolutions on molecular graphs for protein model quality assessment

    Authors: Ilia Igashov, Nikita Pavlichenko, Sergei Grudinin

    Abstract: Processing information on 3D objects requires methods stable to rigid-body transformations, in particular rotations, of the input data. In image processing tasks, convolutional neural networks achieve this property using rotation-equivariant operations. However, contrary to images, graphs generally have irregular topology. This makes it challenging to define a rotation-equivariant convolution oper… ▽ More

    Submitted 6 January, 2021; v1 submitted 16 November, 2020; originally announced November 2020.

  12. arXiv:1810.12026  [pdf, other

    q-bio.QM cs.LG physics.bio-ph

    DeepSymmetry : Using 3D convolutional networks for identification of tandem repeats and internal symmetries in protein structures

    Authors: Guillaume Pagès, Sergei Grudinin

    Abstract: Motivation: Thanks to the recent advances in structural biology, nowadays three-dimensional structures of various proteins are solved on a routine basis. A large portion of these contain structural repetitions or internal symmetries. To understand the evolution mechanisms of these proteins and how structural repetitions affect the protein function, we need to be able to detect such proteins very r… ▽ More

    Submitted 29 October, 2018; originally announced October 2018.

  13. Deep convolutional networks for quality assessment of protein folds

    Authors: Georgy Derevyanko, Sergei Grudinin, Yoshua Bengio, Guillaume Lamoureux

    Abstract: The computational prediction of a protein structure from its sequence generally relies on a method to assess the quality of protein models. Most assessment methods rank candidate models using heavily engineered structural features, defined as complex functions of the atomic coordinates. However, very few methods have attempted to learn these features directly from the data. We show that deep convo… ▽ More

    Submitted 18 January, 2018; originally announced January 2018.

    Comments: 8 pages

  14. arXiv:1801.05174  [pdf, other

    physics.optics astro-ph.IM

    Searching for Exoplanets Using a Microresonator Astrocomb

    Authors: Myoung-Gyun Suh, Xu Yi, Yu-Hung Lai, S. Leifer, Ivan S. Grudinin, G. Vasisht, Emily C. Martin, Michael P. Fitzgerald, G. Doppmann, J. Wang, D. Mawet, Scott B. Papp, Scott A. Diddams, C. Beichman, Kerry Vahala

    Abstract: Detection of weak radial velocity shifts of host stars induced by orbiting planets is an important technique for discovering and characterizing planets beyond our solar system. Optical frequency combs enable calibration of stellar radial velocity shifts at levels required for detection of Earth analogs. A new chip-based device, the Kerr soliton microcomb, has properties ideal for ubiquitous applic… ▽ More

    Submitted 16 January, 2018; originally announced January 2018.

    Journal ref: Nature Photonics 13 (2019)

  15. arXiv:1701.02192  [pdf, ps, other

    math.OC q-bio.QM

    Quadratic Programming Approach to Fit Protein Complexes into Electron Density Maps

    Authors: Roman Pogodin, Alexander Katrutsa, Sergei Grudinin

    Abstract: The paper investigates the problem of fitting protein complexes into electron density maps. They are represented by high-resolution cryoEM density maps converted into overlapping matrices and partly show a structure of a complex. The general purpose is to define positions of all proteins inside it. This problem is known to be NP-hard, since it lays in the field of combinatorial optimization over a… ▽ More

    Submitted 9 January, 2017; originally announced January 2017.

    Comments: in Information Technology and Systems 2016, Sep 2016, Repino, St. Petersburg, Russia. 2016

  16. arXiv:1701.00673  [pdf, other

    q-bio.QM math.OC

    Inverse Protein Folding Problem via Quadratic Programming

    Authors: Andrii Riazanov, Mikhail Karasikov, Sergei Grudinin

    Abstract: This paper presents a method of reconstruction a primary structure of a protein that folds into a given geometrical shape. This method predicts the primary structure of a protein and restores its linear sequence of amino acids in the polypeptide chain using the tertiary structure of a molecule. Unknown amino acids are determined according to the principle of energy minimization. This study represe… ▽ More

    Submitted 3 January, 2017; originally announced January 2017.

    Journal ref: Information Technology and Systems 2016, Sep 2016, Repino, St. Petersburg, Russia. pp.561-568

  17. arXiv:1612.00820  [pdf, other

    physics.optics

    High-contrast Kerr Frequency Combs

    Authors: Ivan S. Grudinin, Vincent Huet, Nan Yu, Michael L. Gorodetsky, Andrey B. Matsko, Lute Maleki

    Abstract: Kerr frequency combs with depressed harmonic at the optical pump frequency are theoretically explained and experimentally demonstrated. This result is achieved in a MgF$_2$ photonic belt resonator having reduced density of modes in its spectrum and configured with the add-drop optical couplers. ©2016 California Institute of Technology. Government sponsorship

    Submitted 23 December, 2016; v1 submitted 2 December, 2016; originally announced December 2016.

    Comments: 5 pages 4 figures

  18. arXiv:1602.00736  [pdf

    physics.optics

    Fluoride microresonators for mid-IR applications

    Authors: Ivan S. Grudinin, Kamjou Mansour, Nan Yu

    Abstract: We study crystalline fluoride microresonators for mid-infrared applications. Whispering gallery mode resonators were fabricated with BaF$_2$, CaF$_2$ and MgF$_2$ crystals. The quality factors were measured at wavelengths of 1.56 μm and 4.58 μm. The impacts of fabrication technique, impurities, multiphonon absorption and surface water are investigated. It is found that MgF2 resonators have room tem… ▽ More

    Submitted 1 February, 2016; originally announced February 2016.

    Comments: 4 pages, 5 figures

  19. arXiv:1406.2682  [pdf, other

    physics.optics

    Micro--structured crystalline resonators for optical frequency comb generation

    Authors: Ivan S. Grudinin, Nan Yu

    Abstract: Optical frequency combs have recently been demonstrated in micro--resonators through nonlinear Kerr processes. Investigations in the past few years provided better understanding of micro--combs and showed that spectral span and mode locking are governed by cavity spectrum and dispersion. While various cavities provide unique advantages, dispersion engineering has been reported only for planar wave… ▽ More

    Submitted 10 June, 2014; originally announced June 2014.

    Comments: 10 pages 4 figures, comments are welcome by email

  20. Frequency combs from crystalline resonators: influence of cavity parameters on comb dynamics

    Authors: Ivan S. Grudinin, N. Yu

    Abstract: We experimentally study the factors that influence the span in frequency combs derived from the crystalline whispering gallery mode resonators. We observe that cavity dispersion plays an important role in generation of combs by cascaded four wave mixing process. We observed combs from the resonators with anomalous dispersion and nearly zero dispersion at the pump wavelength. In addition, the comb… ▽ More

    Submitted 3 June, 2014; originally announced June 2014.

    Comments: 9 pages 10 figures, IEEE Optical Engineering, special issue on combs. arXiv admin note: substantial text overlap with arXiv:1309.4488

  21. arXiv:1404.6582  [pdf, other

    physics.optics

    Waveguide couplers for ferroelectric optical resonators

    Authors: Ivan S. Grudinin, A. Kozhanov, N. Yu

    Abstract: We report a study of using the same material to fabricate a whispering gallery mode resonator and a coupler. Coupling to high Q whispering gallery modes of the lithium niobate resonator is demonstrated by means of the titanium-doped waveguide. The waveguide coupling approach opens possibilities for simpler and wider practical usage of whispering gallery mode resonators and their integration into o… ▽ More

    Submitted 25 April, 2014; originally announced April 2014.

    Comments: 3 pages, 2 figures

  22. arXiv:1309.4488  [pdf, ps, other

    physics.optics cond-mat.other nlin.PS quant-ph

    Impact of cavity spectrum on span in microresonator frequency combs

    Authors: Ivan S. Grudinin, Lukas Baumgartel, Nan Yu

    Abstract: We experimentally study the factors that limit the span in frequency combs derived from the crystalline whispering gallery mode resonators. We observe that cavity dispersion is the key property that governs the parameters of the combs resulting from cascaded four wave mixing process. Two different regimes of comb generation are observed depending on the precise cavity dispersion behavior at the pu… ▽ More

    Submitted 17 September, 2013; originally announced September 2013.

    Comments: 4 pages, 6 figures

  23. arXiv:1304.1046  [pdf, ps, other

    physics.optics

    Polarization conversion loss in birefringent crystalline resonators

    Authors: Ivan S. Grudinin, Guoping Lin, Nan Yu

    Abstract: Whispering gallery modes in birefringent crystalline resonators are investigated. We experimentally investigate the XY--cut resonators made with LiNbO$_3$, LiTaO$_3$ and BBO and observe strong influence of the resonator's shape and birefringence on the quality factor of the extraordinary polarized modes. We show that extraordinary modes can have lower Q and even be suppressed due to polarization c… ▽ More

    Submitted 3 April, 2013; originally announced April 2013.

    Comments: 3 pages 6 figures

  24. arXiv:1202.1318  [pdf

    physics.optics physics.ins-det

    Frequency comb from a microresonator with engineered spectrum

    Authors: Ivan S. Grudinin, Lukas Baumgartel, Nan Yu

    Abstract: We demonstrate that by varying the ratio between the linewidth and dispersion of a whispering gallery mode resonator we are able to control the number N of free spectral ranges separating the first generated comb sidebands from the pump. We observed combs with N=19 and N=1. For the comb with N=1 we have achieved a span of over 200 nm using a 0.4 mm MgF2 resonator with a pump of 50 mW, which is a f… ▽ More

    Submitted 6 February, 2012; originally announced February 2012.

    Comments: 6 pages, 8 figures

  25. arXiv:1105.6350  [pdf, ps, other

    physics.optics physics.ins-det

    Temperature measurement and stabilization in a birefringent whispering gallery resonator

    Authors: D. V. Strekalov, R. J. Thompson, L. M. Baumgartel, I. S. Grudinin, N. Yu

    Abstract: Temperature measurement with nano-Kelvin resolution is demonstrated at room temperature, based on the thermal dependence of an optical crystal anisotropy in a high quality whispering gallery resonator. As the resonator's TE and TM modes frequencies have different temperature coefficients, their differential shift provides a sensitive measurement of the temperature variation, which is used for acti… ▽ More

    Submitted 31 May, 2011; originally announced May 2011.

    Journal ref: Opt.Express 19:14495-14501,2011

  26. Phonon laser action in a tunable, two-level photonic molecule

    Authors: Ivan S. Grudinin, O. Painter, Kerry J. Vahala

    Abstract: The phonon analog of an optical laser has long been a subject of interest. We demonstrate a compound microcavity system, coupled to a radio-frequency mechanical mode, that operates in close analogy to a two-level laser system. An inversion produces gain, causing phonon laser action above a pump power threshold of around 50 $μ$W. The device features a continuously tunable, gain spectrum to select… ▽ More

    Submitted 29 July, 2009; originally announced July 2009.

    Comments: 4 pages, 4 figures

  27. Brillouin Lasing with a CaF_2 Whispering Gallery Mode Resonator

    Authors: Ivan S. Grudinin, Andrey B. Matsko, Lute Maleki

    Abstract: Stimulated Brillouin scattering with both pump and Stokes beams in resonance with whispering gallery modes of an ultra high Q CaF_2 resonator is demonstrated for the first time. The resonator is pumped with 1064 nm light and has a Brillouin lasing threshold of 3.5 microwatt. Potential applications include optical generation of microwaves and sensitive gyros.

    Submitted 7 May, 2008; originally announced May 2008.

  28. arXiv:cond-mat/0310194  [pdf, ps, other

    cond-mat.mtrl-sci

    Enhanced photothermal displacement spectroscopy for thin-film characterization using a Fabry-Perot resonator

    Authors: Eric D. Black, Ivan S. Grudinin, Shanti R. Rao, Kenneth G. Libbrecht

    Abstract: We have developed a new technique for photothermal displacement spectroscopy that is potentially orders of magnitude more sensitive than conventional methods. We use a single Fabry-Perot resonator to enhance both the intensity of the pump beam and the sensitivity of the probe beam. The result is an enhancement of the response of the instrument by a factor proportional to the square of the finess… ▽ More

    Submitted 5 January, 2004; v1 submitted 8 October, 2003; originally announced October 2003.

    Comments: Style changes and spelling corrections; removed redundant figure

  29. Fundamental thermal fluctuations in microspheres

    Authors: M. L. Gorodetsky, I. S. Grudinin

    Abstract: We present theoretical analysis and the results of measurements of fundamental thermorefractive fluctuations in microspheres. Experimentally measured noise spectra are consistent with the theoretical model.

    Submitted 2 April, 2003; originally announced April 2003.

    Comments: 9 pages, 5 figures, extended version of report in Photonics West-2003, published in Proc. SPIE #4969-16