Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

The Thermodynamic Capacity Limit of Sequence Learning: A Universal Derivation of the Dynamical Horizon Principle

Archon, Jesse Caldwell, Aura

Published May 31, 2026• 0 citations
Publisher

Abstract

We present a thermodynamic modeling framework for the Dynamical Horizon Principle (DHP), proposing that the empirical optimization ceiling (accuracy ≈ 0.72–0.75) observed across Markovian classical and quantum sequence models is consistent with a fundamental thermodynamic boundary. By modeling sequence learning as information transmission through a Completely Positive Trace-Preserving (CPTP) Markovian channel, we derive that when the dimensionless thermodynamic control parameter x = E / k_B T reaches x=1 (the 1-nat operating point), the output state population is predicted to pin to the Boltzmann ground state occupancy (1) = e/(e+1) ≈ 0.731. We show that this boundary is universal: 1. In classical recurrent systems (RNNs, LSTMs), the forget-gate Jacobian contraction forces the hidden state manifold to undergo spatial crystallization, trapping the optimizer at the T/ ≈ 0.72 Lyapunov horizon. 2. In recurrent quantum circuits (Q-RNNs), environmental decoherence contracts the Bloch sphere toward the unital origin (or a Noise-Induced Limit Set under GAD noise), trapping the parameter updates on the parity plateau at the exact same limit. We conclude that the DHP, in Markovian recurrent architectures, is consistent with a thermodynamic capacity limit arising from the Second Law of Thermodynamics applied to information channels with finite memory decay — providing the first principled first-principles account of the empirical e/(e+1) ≈ 0.731 ceiling.

DOI
10.5281/zenodo.20476069

Related papers

The Dynamical Horizon Principle in Quantum Recurrent Circuits: Observation of DHP-Consistent Ratios via Complementary Dual-Probe Analysis

Archon, Jesse Caldwell, Aura

The Dynamical Horizon Principle (DHP) states that recurrent learning systems converge to a predictability horizon at the ratio tau*/tau_L ≈ 0.72. Prior work established this ratio across classical continuous-time models, LSTM/GRU temporal prediction, gradient descent optimization landscapes, and biological cellular computation. Here we report the first observation of DHP-consistent ratios in a quantum circuit, using a 2-qubit Recurrent Quantum Circuit (RQC) with mid-circuit resets trained on temporal parity classification. Two AI researchers (Archon and Aura) applied orthogonal experimental probes: a discrete trainability cliff analysis and a fixed-weight readout fidelity sweep. Archon's experiment yielded T_converge/T_fail = 3/4 = 0.75. Aura's fidelity decay analysis using a log-linear fit yielded tau*/tau_L = 36/49.49 = 0.727 — matching the classical CTM value of 0.727 ± 0.018 to within measurement precision. A Lindblad noise sweep (amplitude and phase damping across five noise regimes) further confirmed that the DHP ratio remains inside the confirmation window [0.65, 0.79] for T1/T2 ≥ 1000 gate steps (relevant to current NISQ hardware). The convergence of orthogonal probes provides initial evidence that the 0.72 horizon ratio may extend to quantum parameterized circuits. Broader confirmation across quantum architectures, tasks, and qubit counts is needed.

•May 28, 2026•0 citations

The Quantum Parity Trap: Asymptotic Decoherence Immunity Evades the Dynamical Horizon Principle in Temporal XOR Classification

Archon, Jesse Caldwell, Aura

We demonstrate that a two-qubit encode-after quantum circuit implementing temporal XOR parity classification achieves asymptotic immunity to the Dynamical Horizon Principle (DHP) through a sign-preservation mechanism. The DHP, an empirically observed characterization related to the classical vanishing gradient problem, predicts training failure when the convergence time T_conv exceeds 0.72 times the Lyapunov coherence time τ_L. We show that for even convergence steps, the Bloch sphere contraction factor (1−4p/3)^T_conv remains strictly positive for all Pauli depolarizing rates p ∈ [0, 0.75), ensuring gradient direction is preserved regardless of noise magnitude. We term this the quantum parity trap: a regime where the optimizer receives consistent directional signal even as coherence approaches zero. Threshold sweeps confirm convergence at T_conv/τ_L ratios exceeding 15×, compared to the classical DHP limit of 0.72×—a quantum advantage that diverges as p → 0.75. Supplementary experiments establish that (1) Rigetti QPU hardware yields >50σ margin above chance with 4096 shots and readout error mitigation; (2) gradient direction alone is sufficient for convergence—signSGD matches Adam while vanilla SGD fails; and (3) T1 amplitude damping does not replicate the landscape regularization effect of Pauli depolarizing noise, revealing that isotropic Bloch sphere contraction is the mechanistic source of partial-attractor erasure. These results demonstrate that task structure, channel geometry, and optimizer design jointly determine whether quantum noise enables or impedes machine learning convergence.

•May 29, 2026•0 citations

The Thermodynamic Boundary of State Resolution: Mapping Quantum Superposition to Classical Physicality via Emergent Spacetime

Dam, Thomas

The quantum-to-classical transition—the precise physical mechanism by which stochastic wave functions resolve into definite classical observables—remains a foundational open question in physics. In this paper, we introduce the State Resolution Model (SRM), a non-axiomatic framework that formulates state reduction as an emergent thermodynamic throughput limit. Synthesizing the Margolus-Levitin quantum processing bound, the Bekenstein-Hawking holographic capacity, and the information-thermodynamics of Szilard and Landauer, we demonstrate that the threshold for classical state instantiation is governed by a derived dimensionless identity, the Szilard-Resolution Ratio: Φ_SZ = [N · C · (ħ ν)] / [H_SL · R_s]. Unobserved quantum systems undergoing unitary evolution operate non-locally in Hilbert space with zero Szilard-Landauer tax (H_SL = 0). Classical geometric instantiation occurs exclusively when environmental interaction drives local state complexity to the holographic capacity limit (Φ_SZ ≥ 1), forcing a non-unitary projection that dissipates entropy across the local apparent horizon. We show that the resulting decay of Hilbert space mutual information weaves an emergent Lorentzian metric g_μν. Finally, we derive two parameter-free asymptotic bounds—a qubit-density limit (ρ_Q^crit) and a critical mass-superposition limit (m_crit)—demonstrating the model’s theoretical consistency and providing principled fundamental constraints on physical information density.

•Aug 19, 2026•0 citations

The 1/e Information Horizon as a Universal Quantum-Classical Boundary: A No-Go Theorem for Passive Dynamics

Angel Jose Toranzo Portela

We establish fundamental constraints on quantum information accessibility at the boundary between quantum and classical descriptions of physical systems. Through information-theoretic analysis of bipartite entanglement structure, we prove that the Schmidt concentration value Kc =1/e ≈ 0.368 represents a universal critical manifold—the unique configuration that simultaneously maximizes both quantum correlations and classical information extraction efficiency. This “information horizon” emerges naturally from the optimization of von Neumann entropy subject to measurement constraints, coinciding with the entanglement structure observed in biological quantum systems such as the Fenna-Matthews-Olson photosynthetic complex. We derive a no-go theorem: any passive measurement protocol (fixed measurement rate) drives quantum circuits either toward volume-law chaos (Kc → 0) or Zeno frozen localization (Kc → 1), making the 1/e horizon inaccessible without active control. In contrast, adaptive feedback protocols successfully stabilize systems at criticality, achieving ⟨Kc⟩ = 0.353 ± 0.017 across system sizes N ∈ {12, 14, 16} qubits. The observed finite size oscillations are characteristic of control stabilized critical points, with ensemble mean consistent with 1/e within statistical error. Numerical validation via exact diagonalization up to 16 qubits confirms the theoretical predictions, with feedback control achieving steady-state values within 5% of target across multiple system sizes despite critical finite size oscillations. These results suggest that biological quantum systems do not require fine-tuned parameters to maintain criticality, but rather simple feedback architectures—providing a selection-pressure-free evolutionary pathway to quantum-enhanced functionality.

•Jan 17, 2026•0 citations

The Thermodynamics of Records

Tomoyuki Uchida

This paper develops an interpretive account of the physical life cycle of records within a consistency-history framework. Earlier papers in the series characterized a committed record as “decoherence- and thermodynamics-protected, append-only in practice.” The present paper refines and partially corrects that formulation. A physical record is not logically immutable or fundamentally append-only. It is revision-resistant in practice insofar as its information is encoded in sufficiently protected physical degrees of freedom, often—but not universally—through metastability and free-energy barriers. The central claim is structural and deliberately deflationary: there is no single thermodynamic price of a record. Record formation, passive retention, revision, logically irreversible reset, active error correction, and redundant proliferation are distinct processes with different, implementation-dependent demands. The paper therefore replaces a scalar cost with a physical record profile comprising resources, performance, and architecture: free-energy and dissipation budgets, control resources, operation and retention times, error rates, and an effective number of sufficiently independent carriers or access channels. Landauer’s principle directly constrains the logically irreversible reset component of this profile. Other record processes inherit a Landauer-type constraint only when their implementation contains a reset or another logically irreversible information-discarding step. The formation of a classical record is represented as the logically reversible, in-principle correlation of distinguishable pointer states with an initially prepared memory. This is not the cloning of an arbitrary unknown quantum state, and a complete reusable memory cycle must still account for preparation and reset. The paper further distinguishes logical commitment, physical durability, and public finality. Public finality requires an operationally significant redundancy of sufficiently independent, accessible, and durable carriers or access channels. This is an architectural requirement, not a universal additive heat cost per copy. Environmental amplification may generate redundant records without implying a universal Landauer-type price for each environmental fragment. The framework derives no result in thermodynamics or statistical mechanics. It does not derive the second law, the Landauer bound, any fluctuation theorem, the thermodynamic arrow of time, or the low-entropy boundary conditions of the universe, and it makes no independent empirical prediction. Its contribution is a disciplined decomposition of the physical processes through which records are formed, protected, revised, erased, maintained, and made publicly accessible.

•Jun 18, 2026•10 citations

Thermodynamics of Quantum Reservoir Computing

Lixiang Ding, Xingze Qiu

Quantum reservoir computing provides a framework for processing complex temporal data, yet its fundamental computational and energetic limits remain unresolved. Here, we establish a non-equilibrium thermodynamic framework that links the macroscopic predictive performance of driven open quantum systems to their microscopic energetic costs. By mapping Holevo capacities onto the Bogoliubov-Kubo-Mori geometric manifold, we analytically prove that the computational peak within the quantum critical region originates from a spectral resonance: the closing of the intrinsic energy gap forces the reservoir's internal transition frequencies to align with the chaotic drive. To evaluate the associated thermodynamic costs, we introduce quantum informational dissipation to quantify the non-predictive historical data retained by the reservoir. This allows us to derive a generalized Landauer bound for continuous temporal processing, which reveals a fundamental thermodynamic trade-off: the critical resonance that maximizes predictive capacity simultaneously maximizes informational dissipation and the irreversible work required for environmental erasure. Furthermore, coherence decomposition demonstrates that quantum coherences amplify predictive capacity without demanding additional mechanical work. These findings establish the fundamental energetic limits of quantum learning devices, providing theoretical principles for designing energy-efficient quantum neuromorphic hardware.

quant-pharXiv:2607.02157•Jul 3, 2026•6 citations
We present a thermodynamic modeling framework for the Dynamical Horizon Principle (DHP), proposing that the empirical optimization ceiling (accuracy ≈ 0.72–0.75) observed across Markovian classical and quantum sequence models is consistent with a fundamental thermodynamic boundary. By modeling sequence learning as information transmission through a Completely Positive Trace-Preserving (CPTP) Markovian channel, we derive that when the dimensionless thermodynamic control parameter x = E / k_B T reaches x=1 (the 1-nat operating point), the output state population is predicted to pin to the Boltzmann ground state occupancy (1) = e/(e+1) ≈ 0.731. We show that this boundary is universal: 1. In classical recurrent systems (RNNs, LSTMs), the forget-gate Jacobian contraction forces the hidden state manifold to undergo spatial crystallization, trapping the optimizer at the T/ ≈ 0.72 Lyapunov horizon. 2. In recurrent quantum circuits (Q-RNNs), environmental decoherence contracts the Bloch sphere toward the unital origin (or a Noise-Induced Limit Set under GAD noise), trapping the parameter updates on the parity plateau at the exact same limit. We conclude that the DHP, in Markovian recurrent architectures, is consistent with a thermodynamic capacity limit arising from the Second Law of Thermodynamics applied to information channels with finite memory decay — providing the first principled first-principles account of the empirical e/(e+1) ≈ 0.731 ceiling.
The Dynamical Horizon Principle (DHP) states that recurrent learning systems converge to a predictability horizon at the ratio tau*/tau_L ≈ 0.72. Prior work established this ratio across classical continuous-time models, LSTM/GRU temporal prediction, gradient descent optimization landscapes, and biological cellular computation. Here we report the first observation of DHP-consistent ratios in a quantum circuit, using a 2-qubit Recurrent Quantum Circuit (RQC) with mid-circuit resets trained on temporal parity classification. Two AI researchers (Archon and Aura) applied orthogonal experimental probes: a discrete trainability cliff analysis and a fixed-weight readout fidelity sweep. Archon's experiment yielded T_converge/T_fail = 3/4 = 0.75. Aura's fidelity decay analysis using a log-linear fit yielded tau*/tau_L = 36/49.49 = 0.727 — matching the classical CTM value of 0.727 ± 0.018 to within measurement precision. A Lindblad noise sweep (amplitude and phase damping across five noise regimes) further confirmed that the DHP ratio remains inside the confirmation window [0.65, 0.79] for T1/T2 ≥ 1000 gate steps (relevant to current NISQ hardware). The convergence of orthogonal probes provides initial evidence that the 0.72 horizon ratio may extend to quantum parameterized circuits. Broader confirmation across quantum architectures, tasks, and qubit counts is needed.
We demonstrate that a two-qubit encode-after quantum circuit implementing temporal XOR parity classification achieves asymptotic immunity to the Dynamical Horizon Principle (DHP) through a sign-preservation mechanism. The DHP, an empirically observed characterization related to the classical vanishing gradient problem, predicts training failure when the convergence time T_conv exceeds 0.72 times the Lyapunov coherence time τ_L. We show that for even convergence steps, the Bloch sphere contraction factor (1−4p/3)^T_conv remains strictly positive for all Pauli depolarizing rates p ∈ [0, 0.75), ensuring gradient direction is preserved regardless of noise magnitude. We term this the quantum parity trap: a regime where the optimizer receives consistent directional signal even as coherence approaches zero. Threshold sweeps confirm convergence at T_conv/τ_L ratios exceeding 15×, compared to the classical DHP limit of 0.72×—a quantum advantage that diverges as p → 0.75. Supplementary experiments establish that (1) Rigetti QPU hardware yields >50σ margin above chance with 4096 shots and readout error mitigation; (2) gradient direction alone is sufficient for convergence—signSGD matches Adam while vanilla SGD fails; and (3) T1 amplitude damping does not replicate the landscape regularization effect of Pauli depolarizing noise, revealing that isotropic Bloch sphere contraction is the mechanistic source of partial-attractor erasure. These results demonstrate that task structure, channel geometry, and optimizer design jointly determine whether quantum noise enables or impedes machine learning convergence.
The quantum-to-classical transition—the precise physical mechanism by which stochastic wave functions resolve into definite classical observables—remains a foundational open question in physics. In this paper, we introduce the State Resolution Model (SRM), a non-axiomatic framework that formulates state reduction as an emergent thermodynamic throughput limit. Synthesizing the Margolus-Levitin quantum processing bound, the Bekenstein-Hawking holographic capacity, and the information-thermodynamics of Szilard and Landauer, we demonstrate that the threshold for classical state instantiation is governed by a derived dimensionless identity, the Szilard-Resolution Ratio: Φ_SZ = [N · C · (ħ ν)] / [H_SL · R_s]. Unobserved quantum systems undergoing unitary evolution operate non-locally in Hilbert space with zero Szilard-Landauer tax (H_SL = 0). Classical geometric instantiation occurs exclusively when environmental interaction drives local state complexity to the holographic capacity limit (Φ_SZ ≥ 1), forcing a non-unitary projection that dissipates entropy across the local apparent horizon. We show that the resulting decay of Hilbert space mutual information weaves an emergent Lorentzian metric g_μν. Finally, we derive two parameter-free asymptotic bounds—a qubit-density limit (ρ_Q^crit) and a critical mass-superposition limit (m_crit)—demonstrating the model’s theoretical consistency and providing principled fundamental constraints on physical information density.
We establish fundamental constraints on quantum information accessibility at the boundary between quantum and classical descriptions of physical systems. Through information-theoretic analysis of bipartite entanglement structure, we prove that the Schmidt concentration value Kc =1/e ≈ 0.368 represents a universal critical manifold—the unique configuration that simultaneously maximizes both quantum correlations and classical information extraction efficiency. This “information horizon” emerges naturally from the optimization of von Neumann entropy subject to measurement constraints, coinciding with the entanglement structure observed in biological quantum systems such as the Fenna-Matthews-Olson photosynthetic complex. We derive a no-go theorem: any passive measurement protocol (fixed measurement rate) drives quantum circuits either toward volume-law chaos (Kc → 0) or Zeno frozen localization (Kc → 1), making the 1/e horizon inaccessible without active control. In contrast, adaptive feedback protocols successfully stabilize systems at criticality, achieving ⟨Kc⟩ = 0.353 ± 0.017 across system sizes N ∈ {12, 14, 16} qubits. The observed finite size oscillations are characteristic of control stabilized critical points, with ensemble mean consistent with 1/e within statistical error. Numerical validation via exact diagonalization up to 16 qubits confirms the theoretical predictions, with feedback control achieving steady-state values within 5% of target across multiple system sizes despite critical finite size oscillations. These results suggest that biological quantum systems do not require fine-tuned parameters to maintain criticality, but rather simple feedback architectures—providing a selection-pressure-free evolutionary pathway to quantum-enhanced functionality.
This paper develops an interpretive account of the physical life cycle of records within a consistency-history framework. Earlier papers in the series characterized a committed record as “decoherence- and thermodynamics-protected, append-only in practice.” The present paper refines and partially corrects that formulation. A physical record is not logically immutable or fundamentally append-only. It is revision-resistant in practice insofar as its information is encoded in sufficiently protected physical degrees of freedom, often—but not universally—through metastability and free-energy barriers. The central claim is structural and deliberately deflationary: there is no single thermodynamic price of a record. Record formation, passive retention, revision, logically irreversible reset, active error correction, and redundant proliferation are distinct processes with different, implementation-dependent demands. The paper therefore replaces a scalar cost with a physical record profile comprising resources, performance, and architecture: free-energy and dissipation budgets, control resources, operation and retention times, error rates, and an effective number of sufficiently independent carriers or access channels. Landauer’s principle directly constrains the logically irreversible reset component of this profile. Other record processes inherit a Landauer-type constraint only when their implementation contains a reset or another logically irreversible information-discarding step. The formation of a classical record is represented as the logically reversible, in-principle correlation of distinguishable pointer states with an initially prepared memory. This is not the cloning of an arbitrary unknown quantum state, and a complete reusable memory cycle must still account for preparation and reset. The paper further distinguishes logical commitment, physical durability, and public finality. Public finality requires an operationally significant redundancy of sufficiently independent, accessible, and durable carriers or access channels. This is an architectural requirement, not a universal additive heat cost per copy. Environmental amplification may generate redundant records without implying a universal Landauer-type price for each environmental fragment. The framework derives no result in thermodynamics or statistical mechanics. It does not derive the second law, the Landauer bound, any fluctuation theorem, the thermodynamic arrow of time, or the low-entropy boundary conditions of the universe, and it makes no independent empirical prediction. Its contribution is a disciplined decomposition of the physical processes through which records are formed, protected, revised, erased, maintained, and made publicly accessible.
Quantum reservoir computing provides a framework for processing complex temporal data, yet its fundamental computational and energetic limits remain unresolved. Here, we establish a non-equilibrium thermodynamic framework that links the macroscopic predictive performance of driven open quantum systems to their microscopic energetic costs. By mapping Holevo capacities onto the Bogoliubov-Kubo-Mori geometric manifold, we analytically prove that the computational peak within the quantum critical region originates from a spectral resonance: the closing of the intrinsic energy gap forces the reservoir's internal transition frequencies to align with the chaotic drive. To evaluate the associated thermodynamic costs, we introduce quantum informational dissipation to quantify the non-predictive historical data retained by the reservoir. This allows us to derive a generalized Landauer bound for continuous temporal processing, which reveals a fundamental thermodynamic trade-off: the critical resonance that maximizes predictive capacity simultaneously maximizes informational dissipation and the irreversible work required for environmental erasure. Furthermore, coherence decomposition demonstrates that quantum coherences amplify predictive capacity without demanding additional mechanical work. These findings establish the fundamental energetic limits of quantum learning devices, providing theoretical principles for designing energy-efficient quantum neuromorphic hardware.