aiqbench analysis
This research introduces HAMLET, a novel optimization framework that models machine learning parameter updates as dissipative Hamiltonian dynamics within a geometric energy landscape. By treating the loss function as potential energy and utilizing position-dependent metrics derived from data curvature, the method achieves superior performance on MNIST and CIFAR-10 benchmarks compared to standard optimizers. The work bridges advanced mathematical physics concepts with practical deep learning optimization, offering new insights into trajectory control in high-dimensional parameter spaces.
Hamiltonian Metric Learning and Energy-Based Training: A Dissipative Geometric Framework for Optimization
arXiv:2610.04969v1 Announce Type: new Abstract: Optimization in machine learning is usually expressed through iterative rules that update model parameters using information from the loss landscape. In this work, we study an alternative viewpoint in which parameter optimization is treated as the evolution of a dissipative dynamical system. The model parameters are regarded as generalized coordinates, the loss acts as a potential energy, and a positive-definite metric defines the local kinetic geometry of parameter space. Starting from a variational formulation, we derive the corresponding Hamiltonian dynamics, the geometric force generated by a position-dependent metric, and a metric-compatible Rayleigh dissipation law. The resulting continuous system satisfies a monotonic energy-dissipation relation, while its discrete form allows the influence of curvature on the optimization trajectory to be studied directly. We illustrate the framework using a controlled CIFAR-10 image-reconstruction problem for which the optimum is known analytically. With the image Hessian used as the metric, the curvature dependence of the quadratic modal dynamics is removed. We then reparameterize the same image as a matrix product state, producing a genuinely position-dependent metric and a nonzero geometric force. Poincar\'e return maps provide a complementary phase-space view of the resulting contraction dynamics. These examples establish HAMLET as a geometric, energy-based framework for studying optimization as dissipative motion in parameter space. On a five-seed MNIST MLP benchmark, HAMLET attains the highest mean test accuracy (97.95\%) and the lowest mean test negative log-likelihood (0.0696) among the three evaluated optimizers.
Abstract:Optimization in machine learning is usually expressed through iterative rules that update model parameters using information from the loss landscape. In this work, we study an alternative viewpoint in which parameter optimization is treated as the evolution of a dissipative dynamical system. The model parameters are regarded as generalized coordinates, the loss acts as a potential energy, and a positive-definite metric defines the local kinetic geometry of parameter space. Starting from a variational formulation, we derive the corresponding Hamiltonian dynamics, the geometric force generated by a position-dependent metric, and a metric-compatible Rayleigh dissipation law. The resulting continuous system satisfies a monotonic energy-dissipation relation, while its discrete form allows the influence of curvature on the optimization trajectory to be studied directly. We illustrate the framework using a controlled CIFAR-10 image-reconstruction problem for which the optimum is known analytically. With the image Hessian used as the metric, the curvature dependence of the quadratic modal dynamics is removed. We then reparameterize the same image as a matrix product state, producing a genuinely position-dependent metric and a nonzero geometric force. Poincaré return maps provide a complementary phase-space view of the resulting contraction dynamics. These examples establish HAMLET as a geometric, energy-based framework for studying optimization as dissipative motion in parameter space. On a five-seed MNIST MLP benchmark, HAMLET attains the highest mean test accuracy (97.95\%) and the lowest mean test negative log-likelihood (0.0696) among the three evaluated optimizers.
Machine Learning (cs.LG); Optimization and Control (math.OC); Quantum Physics (quant-ph)
arXiv:2610.04969 [cs.LG]
(or arXiv:2610.04969v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.04969 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 335 words
Related coverage
Symplectic Neural Networks for Learning Non-Separable Hamiltonians
arXiv · Aug 5, 2026
Learning Smooth SE(3) Trajectories under Left-Invariant Riemannian Metrics
arXiv · Aug 4, 2026
News
Quantum Computing Report · Jun 7, 2026
An Analytically Trained Variational Surrogate for Quantum Phase Estimation on NISQ Hardware
arXiv · Jul 24, 2026