Thermodynamic Limits of Physical Intelligence

• URL: http://arxiv.org/abs/2602.05463v1
• Date: Thu, 05 Feb 2026 09:12:43 GMT
• Title: Thermodynamic Limits of Physical Intelligence
• Authors: Koichi Takahashi, Yusuke Hayashi,
• Abstract summary: Modern AI systems achieve remarkable capabilities at the cost of substantial energy consumption.<n>We propose two bits-per-joule metrics under explicit accounting conventions to connect intelligence to physical efficiency.<n>We show how a Landauer-scale closed-cycle benchmark for epiplexity acquisition follows as a corollary of a thermodynamic-learning inequality.
• Score: 0.3580891736370874
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Modern AI systems achieve remarkable capabilities at the cost of substantial energy consumption. To connect intelligence to physical efficiency, we propose two complementary bits-per-joule metrics under explicit accounting conventions: (1) Thermodynamic Epiplexity per Joule -- bits of structural information about a theoretical environment-instance variable newly encoded in an agent's internal state per unit measured energy within a stated boundary -- and (2) Empowerment per Joule -- the embodied sensorimotor channel capacity (control information) per expected energetic cost over a fixed horizon. These provide two axes of physical intelligence: recognition (model-building) vs.control (action influence). Drawing on stochastic thermodynamics, we show how a Landauer-scale closed-cycle benchmark for epiplexity acquisition follows as a corollary of a standard thermodynamic-learning inequality under explicit subsystem assumptions, and we clarify how Landauer-scaled costs act as closed-cycle benchmarks under explicit reset/reuse and boundary-closure assumptions; conversely, we give a simple decoupling construction showing that without such assumptions -- and without charging for externally prepared low-entropy resources (e.g.fresh memory) crossing the boundary -- information gain and in-boundary dissipation need not be tightly linked. For empirical settings where the latent structure variable is unavailable, we align the operational notion of epiplexity with compute-bounded MDL epiplexity and recommend reporting MDL-epiplexity / compression-gain surrogates as companions. Finally, we propose a unified efficiency framework that reports both metrics together with a minimal checklist of boundary/energy accounting, coarse-graining/noise, horizon/reset, and cost conventions to reduce ambiguity and support consistent bits-per-joule comparisons, and we sketch connections to energy-adjusted scaling analyses.

Related papers

• Intrinsic-Energy Joint Embedding Predictive Architectures Induce Quasimetric Spaces [0.764671395172401]
  Joint-Embedding Predictive Architectures (JEPAs) aim to learn representations by predicting target embeddings from context embeddings.<n>Quasimetric Reinforcement Learning (QRL) studies goal-conditioned control through directed distance values (cost-to-go) that support reaching goals under asymmetric dynamics.
  arXiv  Detail & Related papers  (2026-02-12T18:30:27Z)
• Equivariant Evidential Deep Learning for Interatomic Potentials [55.6997213490859]
  Uncertainty quantification is critical for assessing the reliability of machine learning interatomic potentials in molecular dynamics simulations.<n>Existing UQ approaches for MLIPs are often limited by high computational cost or suboptimal performance.<n>We propose textitEquivariant Evidential Deep Learning for Interatomic Potentials ($texte2$IP), a backbone-agnostic framework that models atomic forces and their uncertainty jointly.
  arXiv  Detail & Related papers  (2026-02-11T02:00:25Z)
• BEDS : Bayesian Emergent Dissipative Structures : A Formal Framework for Continuous Inference Under Energy Constraints [0.6345523830122167]
  We introduce BEDS, a formal framework for analyzing inference systems that must maintain beliefs continuously under energy constraints.<n>We prove a central result linking energy, precision, and dissipation.<n>We propose the Gdel-Landauer-Prigogine conjecture, suggesting that closure pathologies across formal systems, computation, and thermodynamics share a common structure.
  arXiv  Detail & Related papers  (2026-01-05T18:21:02Z)
• Going beyond Landauer: Information-cost relations from inference based on the maximum entropy principle [10.21983319647914]
  We derive general information-cost trade-off relations based on the maximum entropy principle.<n>We numerically validate our information-cost trade-off relations using a coupled-qubit system exchanging energy and excitations.<n>Our results underscore the broad utility of maximum-entropy inference in constraining thermodynamic costs for generic finite-time quantum processes.
  arXiv  Detail & Related papers  (2025-09-21T12:38:17Z)
• Hierarchy of Qubit Dynamical Maps in the Presence of Symmetry and Coherence [0.0]
  We prove that U(1) conservation constrains quantum thermodynamic operations through charge conservation of Pauli strings.<n>Our no-go theorem shows that U(1) dynamics cannot generate local coherence from diagonal thermal states.<n>We demonstrate measurable thermodynamic advantages in work extraction and state distinguishability.
  arXiv  Detail & Related papers  (2025-09-05T04:04:34Z)
• Demonstration of energy extraction gain from non-classical correlations [62.615368802619116]
  We show that entanglement governs the amount of extractable energy in a controllable setting. By quantifying both the concurrence of the two-qubit resource state and the energy extraction gain from applying the feedback policy, we corroborate the connection between information and energy.
  arXiv  Detail & Related papers  (2024-04-23T08:44:07Z)
• Disentangled Representation Learning with Transmitted Information Bottleneck [57.22757813140418]
  We present textbfDisTIB (textbfTransmitted textbfInformation textbfBottleneck for textbfDisd representation learning), a novel objective that navigates the balance between information compression and preservation.
  arXiv  Detail & Related papers  (2023-11-03T03:18:40Z)
• A pathway to accurate potential energy curves on NISQ devices [0.483420384410068]
  We present a practical workflow to compute the potential energy curve of a hydrogen molecule on quantum devices. The proposed approach uses an extrapolation scheme to deliver, with only few qubits, full configuration interaction results close to the basis-set limit. We show that despite the limitations imposed by the noisy nature of simulated quantum hardware, it is possible to recover realistic electronic correlation values.
  arXiv  Detail & Related papers  (2022-09-19T11:32:43Z)
• Mixed-Integer Nonlinear Programming for State-based Non-Intrusive Load Monitoring [2.2237337682863125]
  Non-Intrusive Load Monitoring (NILM) is the task of inferring the energy consumption of each appliance given the aggregate signal recorded by a single smart meter. We propose a novel two-stage optimization-based approach for energy disaggregation.
  arXiv  Detail & Related papers  (2021-06-16T22:16:00Z)
• Out-of-time-order correlations and the fine structure of eigenstate thermalisation [58.720142291102135]
  Out-of-time-orderors (OTOCs) have become established as a tool to characterise quantum information dynamics and thermalisation. We show explicitly that the OTOC is indeed a precise tool to explore the fine details of the Eigenstate Thermalisation Hypothesis (ETH) We provide an estimation of the finite-size scaling of $omega_textrmGOE$ for the general class of observables composed of sums of local operators in the infinite-temperature regime.
  arXiv  Detail & Related papers  (2021-03-01T17:51:46Z)
• Risk-Aware Energy Scheduling for Edge Computing with Microgrid: A Multi-Agent Deep Reinforcement Learning Approach [82.6692222294594]
  We study a risk-aware energy scheduling problem for a microgrid-powered MEC network. We derive the solution by applying a multi-agent deep reinforcement learning (MADRL)-based advantage actor-critic (A3C) algorithm with shared neural networks.
  arXiv  Detail & Related papers  (2020-02-21T02:14:38Z)
• Multi-Agent Meta-Reinforcement Learning for Self-Powered and Sustainable Edge Computing Systems [87.4519172058185]
  An effective energy dispatch mechanism for self-powered wireless networks with edge computing capabilities is studied. A novel multi-agent meta-reinforcement learning (MAMRL) framework is proposed to solve the formulated problem. Experimental results show that the proposed MAMRL model can reduce up to 11% non-renewable energy usage and by 22.4% the energy cost.
  arXiv  Detail & Related papers  (2020-02-20T04:58:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.