-
CayleyPy RL: Pathfinding and Reinforcement Learning on Cayley Graphs
Authors:
A. Chervov,
A. Soibelman,
S. Lytkin,
I. Kiselev,
S. Fironov,
A. Lukyanenko,
A. Dolgorukova,
A. Ogurtsov,
F. Petrov,
S. Krymskii,
M. Evseev,
L. Grunvald,
D. Gorodkov,
G. Antiufeev,
G. Verbii,
V. Zamkovoy,
L. Cheldieva,
I. Koltsov,
A. Sychev,
M. Obozov,
A. Eliseev,
S. Nikolenko,
N. Narynbaev,
R. Turtayev,
N. Rokotyan
, et al. (9 additional authors not shown)
Abstract:
This paper is the second in a series of studies on developing efficient artificial intelligence-based approaches to pathfinding on extremely large graphs (e.g. $10^{70}$ nodes) with a focus on Cayley graphs and mathematical applications. The open-source CayleyPy project is a central component of our research. The present paper proposes a novel combination of a reinforcement learning approach with…
▽ More
This paper is the second in a series of studies on developing efficient artificial intelligence-based approaches to pathfinding on extremely large graphs (e.g. $10^{70}$ nodes) with a focus on Cayley graphs and mathematical applications. The open-source CayleyPy project is a central component of our research. The present paper proposes a novel combination of a reinforcement learning approach with a more direct diffusion distance approach from the first paper. Our analysis includes benchmarking various choices for the key building blocks of the approach: architectures of the neural network, generators for the random walks and beam search pathfinding. We compared these methods against the classical computer algebra system GAP, demonstrating that they "overcome the GAP" for the considered examples. As a particular mathematical application we examine the Cayley graph of the symmetric group with cyclic shift and transposition generators. We provide strong support for the OEIS-A186783 conjecture that the diameter is equal to n(n-1)/2 by machine learning and mathematical methods. We identify the conjectured longest element and generate its decomposition of the desired length. We prove a diameter lower bound of n(n-1)/2-n/2 and an upper bound of n(n-1)/2+ 3n by presenting the algorithm with given complexity. We also present several conjectures motivated by numerical experiments, including observations on the central limit phenomenon (with growth approximated by a Gumbel distribution), the uniform distribution for the spectrum of the graph, and a numerical study of sorting networks. To stimulate crowdsourcing activity, we create challenges on the Kaggle platform and invite contributions to improve and benchmark approaches on Cayley graph pathfinding and other tasks.
△ Less
Submitted 25 February, 2025;
originally announced February 2025.
-
A Machine Learning Approach That Beats Large Rubik's Cubes
Authors:
Alexander Chervov,
Kirill Khoruzhii,
Nikita Bukhal,
Jalal Naghiyev,
Vladislav Zamkovoy,
Ivan Koltsov,
Lyudmila Cheldieva,
Arsenii Sychev,
Arsenii Lenin,
Mark Obozov,
Egor Urvanov,
Alexey Romanov
Abstract:
The paper proposes a novel machine learning-based approach to the pathfinding problem on extremely large graphs. This method leverages diffusion distance estimation via a neural network and uses beam search for pathfinding. We demonstrate its efficiency by finding solutions for 4x4x4 and 5x5x5 Rubik's cubes with unprecedentedly short solution lengths, outperforming all available solvers and introd…
▽ More
The paper proposes a novel machine learning-based approach to the pathfinding problem on extremely large graphs. This method leverages diffusion distance estimation via a neural network and uses beam search for pathfinding. We demonstrate its efficiency by finding solutions for 4x4x4 and 5x5x5 Rubik's cubes with unprecedentedly short solution lengths, outperforming all available solvers and introducing the first machine learning solver beyond the 3x3x3 case. In particular, it surpasses every single case of the combined best results in the Kaggle Santa 2023 challenge, which involved over 1,000 teams. For the 3x3x3 Rubik's cube, our approach achieves an optimality rate exceeding 98%, matching the performance of task-specific solvers and significantly outperforming prior solutions such as DeepCubeA (60.3%) and EfficientCube (69.6%). Additionally, our solution is more than 26 times faster in solving 3x3x3 Rubik's cubes while requiring up to 18.5 times less model training time than the most efficient state-of-the-art competitor.
△ Less
Submitted 18 February, 2025;
originally announced February 2025.
-
Moment of inertia and supervortical temperature of gluon plasma
Authors:
Victor V. Braguta,
Maxim N. Chernodub,
Ilya E. Kudrov,
Artem A. Roenko,
Dmitrii A. Sychev
Abstract:
Using lattice simulations, we analyze the influence of uniform rotation on the equation of state of gluodynamics. For a sufficiently slow rotation, the free energy of the system can be expanded into a series of powers of angular velocity. We calculate the moment of inertia given by the quadratic coefficient of this expansion using both analytic continuation and derivative methods, which demonstrat…
▽ More
Using lattice simulations, we analyze the influence of uniform rotation on the equation of state of gluodynamics. For a sufficiently slow rotation, the free energy of the system can be expanded into a series of powers of angular velocity. We calculate the moment of inertia given by the quadratic coefficient of this expansion using both analytic continuation and derivative methods, which demonstrate a good agreement between the results. We find that the moment of inertia unexpectedly takes a negative value below the ``supervortical temperature'' $T_s = 1.50(10) T_c$, vanishes at $T = T_s$, and becomes a positive quantity at higher temperatures. We discuss how our results are related to the scale anomaly and the magnetic gluon condensate. We point out that the negativity of the moment of inertia is in qualitative agreement with our previous lattice calculations, indicating that the rigid rotation increases the critical temperatures in gluodynamics and QCD.
△ Less
Submitted 7 November, 2023;
originally announced November 2023.
-
Negative Barnett effect, negative moment of inertia of gluon plasma and thermal evaporation of chromomagnetic condensate
Authors:
Victor V. Braguta,
Maxim N. Chernodub,
Ilya E. Kudrov,
Artem A. Roenko,
Dmitrii A. Sychev
Abstract:
We discuss the negativity of the moment of inertia of (quark-)gluon plasma in a window of "supervortical" range of temperatures above the deconfining phase transition, $T \simeq (1\dots 1.5) T_c $ found recently in numerical Monte Carlo simulations by two independent methods. In our work, we confirm numerically that the origin of this effect is rooted in the thermal evaporation of the non-perturba…
▽ More
We discuss the negativity of the moment of inertia of (quark-)gluon plasma in a window of "supervortical" range of temperatures above the deconfining phase transition, $T \simeq (1\dots 1.5) T_c $ found recently in numerical Monte Carlo simulations by two independent methods. In our work, we confirm numerically that the origin of this effect is rooted in the thermal evaporation of the non-perturbative chromomagnetic condensate. We argue that the negative moment of inertia of gluon plasma indicates the presence of a novel effect, the negative spin-vortical coupling for gluons resulting in a negative gluonic Barnett effect: the spin polarization of gluons exceeds the total angular momentum of rotating plasma, thus forcing the orbital angular momentum to take negative values in the supervortical range of temperatures.
△ Less
Submitted 24 July, 2024; v1 submitted 24 October, 2023;
originally announced October 2023.
-
Negative moment of inertia and rotational instability of gluon plasma
Authors:
Victor V. Braguta,
Maxim N. Chernodub,
Artem A. Roenko,
Dmitrii A. Sychev
Abstract:
Using first-principle numerical simulations of the lattice SU(3) gauge theory, we calculate the isothermal moment of inertia of the rigidly rotating gluon plasma. We find that the moment of inertia unexpectedly takes a negative value below the "supervortical temperature" $T_s = 1.50(10) T_c$, vanishes at $T = T_s$, and becomes a positive quantity at higher temperatures. The negative moment of iner…
▽ More
Using first-principle numerical simulations of the lattice SU(3) gauge theory, we calculate the isothermal moment of inertia of the rigidly rotating gluon plasma. We find that the moment of inertia unexpectedly takes a negative value below the "supervortical temperature" $T_s = 1.50(10) T_c$, vanishes at $T = T_s$, and becomes a positive quantity at higher temperatures. The negative moment of inertia indicates a thermodynamic instability of rigid rotation. We derive the condition of thermodynamic stability of the vortical plasma and show how it relates to the scale anomaly and the magnetic gluon condensate. The rotational instability of gluon plasma shares a striking similarity with the rotational instabilities of spinning Kerr and Myers-Perry black holes.
△ Less
Submitted 22 March, 2024; v1 submitted 6 March, 2023;
originally announced March 2023.
-
Thermal phase transitions in rotating QCD with dynamical quarks
Authors:
V. V. Braguta,
A. Yu. Kotov,
A. A. Roenko,
D. A. Sychev
Abstract:
Relativistic rotation causes a change of QCD critical temperatures. Various phenomenological and effective models predict a decrease of the critical temperatures in rotating QCD. Nevertheless, lattice simulations showed that the critical temperature in gluodynamics increases due to rotation. We extend the lattice study to the theory with dynamical fermions. We present the first lattice results for…
▽ More
Relativistic rotation causes a change of QCD critical temperatures. Various phenomenological and effective models predict a decrease of the critical temperatures in rotating QCD. Nevertheless, lattice simulations showed that the critical temperature in gluodynamics increases due to rotation. We extend the lattice study to the theory with dynamical fermions. We present the first lattice results for rotating QCD with $N_f=2$ dynamical clover-improved Wilson quarks. We also study separately the effect of rotation on gluonic and fermionic degrees of freedom. It is shown that separate rotations of gluons and fermions have opposite effects on the critical temperatures. In aggregate, the pseudo-critical temperatures in QCD increase with angular velocity. Dependence of the results on the pion mass is also discussed.
△ Less
Submitted 6 December, 2022;
originally announced December 2022.
