publications

2024

1. Accurate Learning of Equivariant Quantum Systems from a Single Ground State

   Štěpán Šmíd ,  and  Roberto Bondesan

   2024
2. Quantum Approximate Optimisation for Not-All-Equal SAT

   Andrew El-Kadi ,  and  Roberto Bondesan

   2024
3. Efficient Learning of Long-Range and Equivariant Quantum Systems

   Štěpán Šmíd ,  and  Roberto Bondesan

   2024

2023

1. The END: An Equivariant Neural Decoder for Quantum Error Correction

   Evgenii Egorov* ,  Roberto Bondesan* ,  and  Max Welling

   In ICLR 2023 Workshop on Physics for Machine Learning , 2023
2. Robust Scheduling with GFlowNets

   David W Zhang ,  Corrado Rainone ,  Markus Peschl , and 1 more author

   In The Eleventh International Conference on Learning Representations , 2023
3. Neural Simulated Annealing

   Alvaro Correia ,  Daniel Worrall ,  and  Roberto Bondesan

   In Artificial Intelligence and Statistics 2023 , 2023

2022

1. Learning Lattice Quantum Field Theories with Equivariant Continuous Flows

   Mathis Gerdes ,  Pim Haan ,  Corrado Rainone , and 2 more authors

   2022
2. Neural Topological Ordering for Computation Graphs

   Mukul Gagrani ,  Corrado Rainone ,  Yang Yang , and 7 more authors

   Advances in Neural Information Processing Systems, 2022
3. Bayesian Optimization for Macro Placement

   Changyong Oh ,  Roberto Bondesan ,  Dana Kianfar , and 6 more authors

   In ICML2022 Workshop on Adaptive Experimental Design and Active Learning in the Real World , 2022
4. Batch Bayesian Optimization on Permutations using the Acquisition Weighted Kernel

   Changyong Oh ,  Roberto Bondesan ,  Efstratios Gavves , and 1 more author

   In Advances in Neural Information Processing Systems , 2022

2021

1. Scaling Up Machine Learning For Quantum Field Theory with Equivariant Continuous Flows

   Pim Haan ,  Corrado Rainone ,  Miranda Cheng , and 1 more author

   In NeurIPS 2021 Workshop on Machine Learning and the Physical Sciences , 2021
2. Deterministic Gibbs Sampling via Ordinary Differential Equations

   Kirill Neklyudov ,  Roberto Bondesan ,  and  Max Welling

   2021
3. Quantum inspired convolutional kernels for convolutional neural networks

   Roberto Bondesan ,  and  Max Welling

   Mar 2021

   US Patent App. 17/031,501
4. The Hintons in your Neural Network: a Quantum Field Theory View of Deep Learning

   Roberto Bondesan ,  and  Max Welling

   In Proceedings of the 38th International Conference on Machine Learning , Mar 2021

   Abs PDF

   In this work we develop a quantum field theory formalism for deep learning, where input signals are encoded in Gaussian states, a generalization of Gaussian processes which encode the agent’s uncertainty about the input signal. We show how to represent linear and non-linear layers as unitary quantum gates, and interpret the fundamental excitations of the quantum model as particles, dubbed “Hintons”. On top of opening a new perspective and techniques for studying neural networks, the quantum formulation is well suited for optical quantum computing, and provides quantum deformations of neural networks that can be run efficiently on those devices. Finally, we discuss a semi-classical limit of the quantum deformed models which is amenable to classical simulation.

5. Probabilistic Numeric Convolutional Neural Networks

   Marc Anton Finzi ,  Roberto Bondesan ,  and  Max Welling

   In International Conference on Learning Representations , Mar 2021

2020

1. Quantum Deformed Neural Networks

   Roberto Bondesan ,  and  Max Welling

   Mar 2020

2019

1. Learning symmetries of classical integrable systems

   Roberto Bondesan ,  and  Austen Lamacraft

   In ICML 2019 Workshop on Theoretical Physics for Deep Learning , Mar 2019

2018

1. Effective edge state dynamics in the fractional quantum Hall effect

   R. Fern ,  R. Bondesan ,  and  S. H. Simon

   Phys. Rev. B, Mar 2018
2. Structure of edge-state inner products in the fractional quantum Hall effect

   R. Fern ,  R. Bondesan ,  and  S. H. Simon

   Phys. Rev. B, Mar 2018

2017

1. Classical topological paramagnetism

   R. Bondesan ,  and  Z. Ringel

   Phys. Rev. B, May 2017
2. Gaussian free fields at the integer quantum Hall plateau transition

   R. Bondesan ,  D. Wieczorek ,  and  M.R. Zirnbauer

   Nuclear Physics B, May 2017
3. Critical behaviour of spanning forests on random planar graphs

   Roberto Bondesan ,  Sergio Caracciolo ,  and  Andrea Sportiello

   Journal of Physics A: Mathematical and Theoretical, Jan 2017

2016

1. Infinite Dimensional Matrix Product States for Long-Range Quantum Spin Models

   Roberto Bondesan ,  and  Thomas Quella

   In Lie Theory and Its Applications in Physics , Jan 2016

2015

1. Chiral SU(2)k currents as local operators in vertex models and spin chains

   R Bondesan ,  J Dubail ,  A Faribault , and 1 more author

   Journal of Physics A: Mathematical and Theoretical, Jan 2015

2014

1. Infinite matrix product states for long-range SU(N) spin models

   Roberto Bondesan ,  and  Thomas Quella

   Nuclear Physics B, Jan 2014

   Abs

   We construct 1D and 2D long-range SU(N) spin models as parent Hamiltonians associated with infinite matrix product states. The latter are constructed from correlators of primary fields in the SU(N)1 WZW model. Since the resulting groundstates are of Gutzwiller–Jastrow type, our models can be regarded as lattice discretizations of fractional quantum Hall systems. We then focus on two specific types of 1D spin chains with spins located on the unit circle, a uniform and an alternating arrangement. For an equidistant distribution of identical spins we establish an explicit connection to the SU(N) Haldane–Shastry model, thereby proving that the model is critical and described by a SU(N)1 WZW model. In contrast, while turning out to be critical as well, the alternating model can only be treated numerically. Our numerical results rely on a reformulation of the original problem in terms of loop models.

2. Pure Scaling Operators at the Integer Quantum Hall Plateau Transition

   R. Bondesan ,  D. Wieczorek ,  and  M. R. Zirnbauer

   Phys. Rev. Lett., May 2014

2013

1. Topological and symmetry broken phases of ZN parafermions in one dimension

   Roberto Bondesan ,  and  Thomas Quella

   Journal of Statistical Mechanics: Theory and Experiment, Oct 2013
2. Rectangular amplitudes, conformal blocks, and applications to loop models

   Roberto Bondesan ,  Jesper L. Jacobsen ,  and  Hubert Saleur

   Nuclear Physics B, Oct 2013

   Abs

   In this paper we continue the investigation of partition functions of critical systems on a rectangle initiated in [R. Bondesan, et al., Nucl. Phys. B 862 (2012) 553–575]. Here we develop a general formalism of rectangle boundary states using conformal field theory, adapted to describe geometries supporting different boundary conditions. We discuss the computation of rectangular amplitudes and their modular properties, presenting explicit results for the case of free theories. In a second part of the paper we focus on applications to loop models, discussing in details lattice discretizations using both numerical and analytical calculations. These results allow to interpret geometrically conformal blocks, and as an application we derive new probability formulas for self-avoiding walks.

2012

1. Conformal boundary state for the rectangular geometry

   R. Bondesan ,  J. Dubail ,  J.L. Jacobsen , and 1 more author

   Nuclear Physics B, Oct 2012

   Abs

   We discuss conformal field theories (CFTs) in rectangular geometries, and develop a formalism that involves a conformal boundary state for the 1+1d open system. We focus on the case of homogeneous boundary conditions (no insertion of a boundary condition changing operator), for which we derive an explicit expression of the associated boundary state, valid for any arbitrary CFT. We check the validity of our solution, comparing it with known results for partition functions, numerical simulations of lattice discretizations, and coherent state expressions for free theories.

2. Exact Exponents for the Spin Quantum Hall Transition in the Presence of Multiple Edge Channels

   R. Bondesan ,  I. A. Gruzberg ,  J. L. Jacobsen , and 2 more authors

   Phys. Rev. Lett., Mar 2012

2011

1. Edge states and conformal boundary conditions in super spin chains and super sigma models

   Roberto Bondesan ,  Jesper L. Jacobsen ,  and  Hubert Saleur

   Nuclear Physics B, Mar 2011

   Abs

   The sigma models on projective superspaces CPN+M−1|N with topological angle θ=πmod2π flow to non-unitary, logarithmic conformal field theories in the low-energy limit. In this paper, we determine the exact spectrum of these theories for all open boundary conditions preserving the full global symmetry of the model, generalizing recent work on the particular case M=0 [C. Candu et al., JHEP 1002 (2010) 015]. In the sigma model setting, these boundary conditions are associated with complex line bundles, and are labelled by an integer, related with the exact value of θ. Our approach relies on a spin chain regularization, where the boundary conditions now correspond to the introduction of additional edge states. The exact values of the exponents then follow from a lengthy algebraic analysis, a reformulation of the spin chain in terms of crossing and non-crossing loops (represented as a certain subalgebra of the Brauer algebra), and earlier results on the so-called one- and two-boundary Temperley–Lieb algebras (also known as blob algebras). A remarkable result is that the exponents, in general, turn out to be irrational. The case M=1 has direct applications to the spin quantum Hall effect, which will be discussed in a sequel.

2008

1. Single-Molecule Observation of Anomalous Electrohydrodynamic Orientation of Microtubules

   M. G. L. Heuvel ,  R. Bondesan ,  M. Cosentino Lagomarsino , and 1 more author

   Phys. Rev. Lett., Sep 2008