The Work Capacity of Channels with Memory: Maximum Extractable Work in Percept-Action Loops

• URL: http://arxiv.org/abs/2504.06209v1
• Date: Tue, 08 Apr 2025 16:54:20 GMT
• Title: The Work Capacity of Channels with Memory: Maximum Extractable Work in Percept-Action Loops
• Authors: Lukas J. Fiderer, Paul C. Barth, Isaac D. Smith, Hans J. Briegel,
• Abstract summary: We analyze the thermodynamics of information processing in percept-action loops.<n>We introduce the concept of work capacity -- the maximum rate at which an agent can expect to extract work from its environment.
• Score: 0.5999777817331317
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Predicting future observations plays a central role in machine learning, biology, economics, and many other fields. It lies at the heart of organizational principles such as the variational free energy principle and has even been shown -- based on the second law of thermodynamics -- to be necessary for reaching the fundamental energetic limits of sequential information processing. While the usefulness of the predictive paradigm is undisputed, complex adaptive systems that interact with their environment are more than just predictive machines: they have the power to act upon their environment and cause change. In this work, we develop a framework to analyze the thermodynamics of information processing in percept-action loops -- a model of agent-environment interaction -- allowing us to investigate the thermodynamic implications of actions and percepts on equal footing. To this end, we introduce the concept of work capacity -- the maximum rate at which an agent can expect to extract work from its environment. Our results reveal that neither of two previously established design principles for work-efficient agents -- maximizing predictive power and forgetting past actions -- remains optimal in environments where actions have observable consequences. Instead, a trade-off emerges: work-efficient agents must balance prediction and forgetting, as remembering past actions can reduce the available free energy. This highlights a fundamental departure from the thermodynamics of passive observation, suggesting that prediction and energy efficiency may be at odds in active learning systems.

Related papers

• Work extractability from energy eigenstates under optimized local operations [0.0]
  We examine the relationship between the second law of thermodynamics and the energy eigenstates of quantum many-body systems. We numerically demonstrate that such a strikingly different behavior can be attributed to the number of athermal energy eigenstates.
  arXiv  Detail & Related papers  (2023-08-07T12:34:09Z)
• Energy Transformer [64.22957136952725]
  Our work combines aspects of three promising paradigms in machine learning, namely, attention mechanism, energy-based models, and associative memory. We propose a novel architecture, called the Energy Transformer (or ET for short), that uses a sequence of attention layers that are purposely designed to minimize a specifically engineered energy function.
  arXiv  Detail & Related papers  (2023-02-14T18:51:22Z)
• A Neural Active Inference Model of Perceptual-Motor Learning [62.39667564455059]
  The active inference framework (AIF) is a promising new computational framework grounded in contemporary neuroscience. In this study, we test the ability for the AIF to capture the role of anticipation in the visual guidance of action in humans. We present a novel formulation of the prior function that maps a multi-dimensional world-state to a uni-dimensional distribution of free-energy.
  arXiv  Detail & Related papers  (2022-11-16T20:00:38Z)
• The Free Energy Principle for Perception and Action: A Deep Learning Perspective [4.6956495676681484]
  The free energy principle, and its corollary active inference, constitute a bio-inspired theory that assumes biological agents act to remain in a restricted set of preferred states of the world. Under this principle, biological agents learn a generative model of the world and plan actions in the future that will maintain the agent in a homeostatic state that satisfies its preferences. This manuscript probes newer perspectives for the active inference framework, grounding its theoretical aspects into more pragmatic affairs.
  arXiv  Detail & Related papers  (2022-07-13T11:07:03Z)
• A-ACT: Action Anticipation through Cycle Transformations [89.83027919085289]
  We take a step back to analyze how the human capability to anticipate the future can be transferred to machine learning algorithms. A recent study on human psychology explains that, in anticipating an occurrence, the human brain counts on both systems. In this work, we study the impact of each system for the task of action anticipation and introduce a paradigm to integrate them in a learning framework.
  arXiv  Detail & Related papers  (2022-04-02T21:50:45Z)
• Emergence of energy-avoiding and energy-seeking behaviours in nonequilibrium dissipative quantum systems [0.0]
  We find some lifelike behaviours and functionalities in elementary dissipative quantum systems driven out of equilibrium. Specifically, we find both energy-avoiding (low steady dissipation) and energy-seeking behaviours (high steady dissipation) We also find emergent functionalities, namely, a self-organized thermal gradient in the system's environment (in the energy-seeking mode) and an active equilibration against thermal gradients (in the energy-avoiding mode)
  arXiv  Detail & Related papers  (2022-01-12T14:54:08Z)
• Class of quasiprobability distributions of work and initial quantum coherence [0.0]
  We study a class of quasiprobability distributions of work, which give an average work equal to the average energy change of the system. We find a fluctuation theorem involving quantum coherence, from which follows a second law of thermodynamics.
  arXiv  Detail & Related papers  (2021-10-03T10:56:07Z)
• Collective effects on the performance and stability of quantum heat engines [62.997667081978825]
  Recent predictions for quantum-mechanical enhancements in the operation of small heat engines have raised renewed interest. One essential question is whether collective effects may help to carry enhancements over larger scales. We study how power, efficiency and constancy scale with the number of spins composing the engine.
  arXiv  Detail & Related papers  (2021-06-25T18:00:07Z)
• Maximal power for heat engines: role of asymmetric interaction times [110.83289076967895]
  We introduce the idea of adjusting the interaction time asymmetry in order to optimize the engine performance. Distinct optimization protocols are analyzed in the framework of thermodynamics.
  arXiv  Detail & Related papers  (2020-12-16T22:26:14Z)
• Thermodynamic Machine Learning through Maximum Work Production [0.0]
  We introduce the thermodynamic principle that work production is the most relevant performance metric for an adaptive physical agent. We show that selecting the maximum-work agent for given environmental data corresponds to finding the maximum-likelihood model. In this way, work emerges as an organizing principle that underlies learning in adaptive thermodynamic systems.
  arXiv  Detail & Related papers  (2020-06-27T17:57:24Z)
• Whence the Expected Free Energy? [68.8204255655161]
  We show that the Expected Free Energy (EFE) is not simply "the free energy in the future" We then develop a novel objective, the Free-Energy of the Expected Future (FEEF)
  arXiv  Detail & Related papers  (2020-04-17T09:06:56Z)

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.