REVIEW 2 major objections 6 minor 53 references
LyLA-Therm: Lyapunov-based Langevin Adaptive Thermodynamic Neural Network Controller
T0 review · 2 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash
Pith's one-line read This paper claims that adding temperature-scaled Langevin noise to a Lyapunov-based deep neural network update law preserves probabilistic ultimate boundedness and improves tracking and approximation error by about 20% in simulation.
desk verdict The LyLA-Therm design is a plausible and genuinely new way to add constructive exploration noise to Lb-DNN control, but the central UUB-p proof is technically unsupported because the diffusion term is not a state-dependent function of the pre-increment state; I agree with the reject verdict. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the LyLA-Therm update law (32): a projected Ito stochastic differential equation d theta_hat = gamma proj(Phi'^T e + 1/2(p+1) gamma kT e^T partial mu/partial theta_hat - sigma theta_hat) dt + gamma proj(sqrt(kT T) d omega), with T = e^T mu. The drift term is the negative gradient of the generalized internal energy (16), which handles exploitation; the diffusion term, scaled by the square root of the user-selected temperature law, handles exploration. The stability argument uses the infinitesimal generator of the Lyapunov function (35) and a UUB-p lemma from the stochastic-control literature; the projection operator (3) keeps theta_hat inside a compact convex set, a
What would settle it
Check whether the diffusion coefficient in (33)-(34) is measurable with respect to the sigma-field generated by states up to time t (pre-increment). If the sign condition deciding Case 1 versus Case 2 uses the same d omega that enters the increment, the coefficient depends on the future increment, so the process is not an Ito diffusion in the usual sense; computing the generator as in (41) and invoking [39, Lemma 1] would then be invalid. A concrete numerical test: simulate (32) with tiny step sizes, estimate the escape probability P(sup ||z|| >= sqrt(lambda/alpha1)), and compare with the boun
Extended reading notes
Core claim
On its own terms, the paper establishes that stochastic exploration can be built into Lyapunov-based DNN adaptive control without sacrificing stability guarantees. The update law (15)/(32) is a projected Ito SDE whose drift is the negative gradient of the generalized internal energy U = e^T e_dot + 1/2 sigma ||theta_hat||^2, and whose diffusion is proj(sqrt(kT T) d omega) with temperature T = e^T mu chosen by the designer. Theorem 1 states that if the initial concatenated error z(0) lies in a set S and a gain condition holds, the closed-loop process is uniformly ultimately bounded in probability: the probability that ||z|| stays below sqrt(lambda/alpha1) is at least 1 - vartheta, with varthe
Load-bearing premise
The proof needs the projected update law with its case-switching diffusion term to be a standard Ito diffusion whose infinitesimal generator can be computed from the pre-increment state; but the diffusion switches based on the sign of the very Brownian increment it contains, so the standard generator identity (41) may not apply as written.
Editorial extensions
If this is right
- Practitioners can add exploratory noise to Lyapunov-based DNN adaptation by choosing a temperature law mu, with the tracking-error bound (40) providing an explicit probabilistic certificate.
- The same Lyapunov machinery accommodates different temperature schedules; the paper tests three (mu = 9e, mu = e(0.01||x||^2 + 9), mu = e(0.01||theta_hat||^2 + 9)) and reports all three preserve the boundedness guarantee.
- Because exploration is part of the update law rather than an external dither signal, the controller can be designed to compensate for the noise's cost through the mu-dependent term in (22).
- If the theorem is correct, the exploration-exploitation trade-off in adaptive DNN control is no longer purely heuristic: the probability of exceeding a desired error bound is quantitatively controlled by vartheta in (39).
Reading between the lines
- The temperature law's dependence on the parameter estimate norm (as in mu4) suggests a general scheduling principle: exploration can be automatically suppressed as the estimate matures, a property that could transfer to other online learning settings beyond control.
- A natural extension is to connect the generalized temperature T to an estimate of approximation uncertainty, so that exploration is concentrated where the DNN is least certain; the paper does not explore this, but the diffusion structure (18) makes it a direct modification.
- If the stochastic-calculus issue with the case-switching diffusion is repaired, the bound (40) could be sharpened into an explicit exploration-exploitation trade-off, showing how much tracking performance must be sacrificed for a given level of parameter-space exploration.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes LyLA-Therm, a stochastic adaptive controller for uncertain nonlinear systems. Parameter estimates evolve according to a Langevin-type SDE (15) with drift chosen to minimize a 'generalized internal energy' and diffusion scaled by a user-selected 'generalized temperature law' T = e^⊤ μ. A feedback controller (22) is added to compensate for part of the stochasticity. The main theoretical claim (Theorem 1) is that the tracking error and parameter estimation error are uniformly ultimately bounded in probability, with the proof based on a stochastic Lyapunov analysis using an infinitesimal generator. Simulations on a five-dimensional system compare the proposed method against a deterministic Lb-DNN baseline and report improvements of roughly 20% in RMS tracking error, 20% in function approximation error, and 6–11% in off-trajectory approximation error.
Significance. If the stability analysis were correct, the paper would make a useful contribution: it gives a constructive, non-heuristic way to add exploration noise to Lyapunov-based DNN adaptation while preserving a probabilistic ultimate boundedness guarantee, and it demonstrates nontrivial empirical improvements over a deterministic baseline. The idea of tying the noise intensity to a decaying 'temperature' that depends on the current error is appealing and is tested with three different temperature laws. However, the central proof rests on a stochastic-calculus step that is invalid as written: the diffusion coefficient in the SDE depends on the sign of the very Brownian increment appearing in the same equation. Because the generator computation is the load-bearing part of Theorem 1, the advertised UUB-p guarantee is currently unsupported. The manuscript also leaves the existence of a solution for the proposed SDE as an unproved assumption rather than a derived property. These are not merely presentation issues; they concern the mathematical core of the claimed result and cannot be repaired by local editing.
major comments (2)
- [Section V, Eq. (34) and Eq. (41)] The infinitesimal-generator computation is invalid. In (33), Case 1 vs Case 2 is selected based on the sign of ∇P^⊤ sqrt(kT T) dω for the same Brownian increment dω that appears in the diffusion term. Therefore G1(z) and G2(z) in (34) are not functions of the pre-increment state z; they depend on the realization of dω itself. The process is not shown to be a standard Ito diffusion of the form dz = F(z)dt + G(z)dω with G(z) a state-dependent matrix. Consequently, the generator formula (5) cannot be applied, and the computations in (41)–(45) do not follow. Lemma 1's assumption A2 (continuous strong Markov process) is also not verified. Since the common bound (43) feeds directly into (50)–(52) and hence into (40), the UUB-p statement in Theorem 1 is unsupported.
- [Section IV-B, Lemma 2 and Eq. (15)] The SDE (15) is not well posed as an Ito equation. The diffusion term proj(ς(x,θ^,t)dω) applies the projection operator to the entire Brownian increment, so the noise coefficient is a nonlinear function of dω, not a measurable function of the state before the increment. Lemma 2's proof argues that the 'diffusion increment' satisfies ∇P^⊤ proj(ςdω) ≤ 0 on the boundary, but this is a pathwise statement about a projected realization, not a property of a state-dependent diffusion coefficient. Thus the boundedness of θ^ and, more importantly, the existence and uniqueness of a strong solution on the interval I are not established. This is a separate but related gap: the proof assumes continuity of z on I with probability one, but the stochastic differential equation used to generate z has not been shown to admit such a solution.
minor comments (6)
- [Section IV-A, Eq. (14)] The claim that 'there exists s which ensures L(θ) is convex' is not justified for a general DNN parameterization; the mapping θ ↦ ||f(x) − Φ(x,θ)||² is typically nonconvex in θ, and adding s||θ||² need not make it strictly convex. The cited [46] may provide conditions, but they are not stated. If strict convexity is not needed for ultimate boundedness (e.g., any minimizer in Ω can be used), this should be clarified.
- [Section V, Theorem 1 proof] The sentence 'Since VL(0) = 0' is false for a generic nonzero initial condition z(0). This appears to be a typo referring to the zero-value assumption in Lemma 1, but it should be corrected.
- [Section V, Eq. (45)] The text says 'Substituting F(z) and G2(z) into (41)', but it should be (44) (the Case 2 generator).
- [Section IV-B, Eq. (33) and proof] The case definitions in (33) and in the proof of Theorem 1 are inconsistent: (33) uses P and the condition θ^ ∈ Πϵ\◦Π, while the proof refers to k(θ) < 0 and k(θ) = 0. The notation (h vs k, P vs ∇P) should be aligned.
- [Section VI] The simulation section reports a single stochastic run for each configuration. Since the update law is stochastic, reporting mean and variance over multiple independent trials would substantially strengthen the empirical comparison; the current single-run percentages are not a reliable measure of improvement.
- [Throughout] There are many typographical and notational slips (e.g., 'parrameter' in reference [23], 'recognizd' in the biography, and missing punctuation in several displayed equations). Careful proofreading is needed.
Circularity Check
No significant circularity; LyLA-Therm's stability claim is a constructive Lyapunov design supported by independent generic lemmas.
full rationale
The paper's central claim—probabilistic UUB of tracking and parameter estimation errors—is derived through a standard constructive Lyapunov argument. The update law (15) and controller (22) are explicitly designed so that the Lyapunov function in (35) satisfies the differential inequality (51), and Theorem 1 then applies a general stochastic UUB lemma. No quantity is fitted to simulation data and then relabeled a prediction; the temperature law T and energy U are user-defined design objects, not empirical inputs. The simulation improvements are direct comparisons against a deterministic baseline, not predictions forced by fitted parameters. The only possible circularity concern is self-citation: Lemma 1 is taken from the authors' prior work [39] and the Taylor-remainder bound in Lemma 3 from [49]. However, both are generic mathematical lemmas with stated assumptions that do not include the target result of this paper, and the core Lyapunov inequality is established from the actual closed-loop error dynamics rather than from those citations. Thus the self-citations are independent support, not circular. The stochastic-calculus concern about the case-dependent diffusion matrix in (34) is a potential correctness gap in the proof, but it is not circularity: it does not reduce the claimed result to its own inputs by definition or by construction.
Assumptions & free parameters
free parameters (5)
- kT =
0.03
- ke =
100
- gamma =
1
- sigma =
0.001
- mu design constants =
9 and 0.01 for S2-S4
assumptions (7)
- domain assumption Activations and their first two derivatives satisfy uniform bounds (Assumption 1, Eq. 8)
- domain assumption Desired trajectory and its derivative are bounded with known constants (Assumption 2)
- domain assumption Unknown drift f can be represented as Φ(x,θ*)+ε(x) over a compact set Ω with bounded reconstruction error ε (Eqs. 11-12)
- ad hoc to paper The loss L is strictly convex over Ω for some regularization constant s, giving a unique θ* (Section IV-A)
- ad hoc to paper z(t) is a continuous strong Markov process on I and Ito's formula can be applied to (34)
- domain assumption Polynomial Lagrange remainder bound from [49, Thm. 1] with known constants
- standard math Smooth projection operator property from [51] and invariance result in Lemma 2
invented entities (2)
-
generalized internal energy U
-
generalized temperature law T=e^Tµ
Cite this review
Pith. "Pith review of LyLA-Therm: Lyapunov-based Langevin Adaptive Thermodynamic Neural Network Controller." pith.science (2026). https://pith.science/paper/HSML7QIA
@misc{pith2026250814989,
author = {Pith},
title = {Pith review of: LyLA-Therm: Lyapunov-based Langevin Adaptive Thermodynamic Neural Network Controller},
year = {2026},
howpublished = {\url{https://pith.science/paper/HSML7QIA}},
note = {Machine review of arXiv:2508.14989}
}
read the original abstract
Thermodynamic principles can be employed to design parameter update laws that address challenges such as the exploration vs. exploitation dilemma. In this paper, inspired by the Langevin equation, an update law is developed for a Lyapunov-based DNN control method, taking the form of a stochastic differential equation. The drift term is designed to minimize the system's generalized internal energy, while the diffusion term is governed by a user-selected generalized temperature law, allowing for more controlled fluctuations. The minimization of generalized internal energy in this design fulfills the exploitation objective, while the temperature-based stochastic noise ensures sufficient exploration. Using a Lyapunov-based stability analysis, the proposed Lyapunov-based Langevin Adaptive Thermodynamic (LyLA-Therm) neural network controller achieves probabilistic convergence of the tracking and parameter estimation errors to an ultimate bound. Simulation results demonstrate the effectiveness of the proposed approach, with the LyLA-Therm architecture achieving up to 20.66% improvement in tracking errors, up to 20.89% improvement in function approximation errors, and up to 11.31% improvement in off-trajectory function approximation errors compared to the baseline deterministic approach.
Reference graph
Works this paper leans on
-
[39]
Lyapunov-Based Deep Neural Networks for Adaptive Control of Stochastic Nonlinear Systems
S. Akbari, C. F. Nino, O. S. Patil, and W. E. Dixon, “Lyapunov- based deep neural networks for adaptive control of stochastic nonlinear systems,” arXiv preprint arXiv:2412.21095 , 2024
work page Pith review arXiv 2024
-
[49]
Bounds on Deep Neural Network Partial Derivatives with Respect to Parameters
O. S. Patil, B. C. Fallin, C. F. Nino, R. G. Hart, and W. E. Dixon, “Bounds on deep neural network partial derivatives with respect to parameters,” arXiv preprint arXiv:2503.21007 , 2025
work page Pith review arXiv 2025
-
[1]
W. M. Haddad, V . Chellaboina, and S. G. Nersesov, Thermodynamics: A dynamical systems approach . Princeton University Press, 2009
work page 2009
-
[2]
W. M. Haddad, A dynamical systems theory of thermodynamics . Prince- ton University Press, 2019
work page 2019
-
[3]
Distributed nonlinear control algorithms for network consensus,
Q. Hui and W. M. Haddad, “Distributed nonlinear control algorithms for network consensus,” Automatica, vol. 44, no. 9, pp. 2375–2381, 2008
work page 2008
-
[4]
Finite-time semistability and consensus for nonlinear dynamical networks,
Q. Hui, W. M. Haddad, and S. P. Bhat, “Finite-time semistability and consensus for nonlinear dynamical networks,” IEEE Trans. on Autom. Control, vol. 53, no. 8, pp. 1887–1900, 2008
work page 1900
-
[5]
Thermodynamics-based control of network systems,
J. M. Berg, D. Maithripala, Q. Hui, and W. M. Haddad, “Thermodynamics-based control of network systems,” Journal of Dy- namic Systems, Measurement, and Control , vol. 135, no. 5, p. 051003, 2013
work page 2013
-
[6]
The role of systems biology, neuroscience, and thermodynamics in network control and learning,
W. M. Haddad, “The role of systems biology, neuroscience, and thermodynamics in network control and learning,” in Handbook of Reinforcement Learning and Control , pp. 763–817, Springer, 2021
work page 2021
Show all 53 references
-
[7]
W. M. Haddad, V . Chellaboina, and S. G. Nersesov, Impulsive and hybrid dynamical systems: stability, dissipativity, and control . Princeton University Press, 2006
2006
-
[8]
Energy-and entropy-based stabilization for lossless dynamical systems via hybrid controllers,
W. M. Haddad, V . Chellaboina, Q. Hui, and S. G. Nersesov, “Energy-and entropy-based stabilization for lossless dynamical systems via hybrid controllers,” IEEE Trans. on Autom. Control , vol. 52, no. 9, pp. 1604– 1614, 2007
2007
-
[9]
Hybrid decentralized maximum entropy control for large-scale dynamical sys- tems,
W. M. Haddad, Q. Hui, V . Chellaboina, and S. G. Nersesov, “Hybrid decentralized maximum entropy control for large-scale dynamical sys- tems,” Nonlinear Anal.: Hybrid Syst. , vol. 1, no. 2, pp. 244–263, 2007
2007
-
[10]
A thermodynamic-based control architecture for semistability and consensus of discrete-time nonlinear network sys- tems,
W. M. Haddad and J. Lee, “A thermodynamic-based control architecture for semistability and consensus of discrete-time nonlinear network sys- tems,” in IEEE Conf. on Control Technolog. & Appl. (CCTA) , pp. 499– 504, 2021
2021
-
[11]
A hybrid thermodynamic control protocol for semistability and consensus of network systems with intermittent information,
W. M. Haddad and M. Chahine, “A hybrid thermodynamic control protocol for semistability and consensus of network systems with intermittent information,” IEEE Trans. on Autom. Control, vol. 66, no. 6, pp. 2754–2761, 2020
2020
-
[12]
Kinetic characterization of heat bath and the energetics of thermal ratchet models,
K. Sekimoto, “Kinetic characterization of heat bath and the energetics of thermal ratchet models,” J. of the Phys. Soc. of Japan , vol. 66, no. 5, pp. 1234–1237, 1997
1997
-
[13]
Langevin equation and thermodynamics,
K. Sekimoto, “Langevin equation and thermodynamics,” Prog. of Theor . Phys. Suppl. , vol. 130, pp. 17–27, 1998
1998
-
[14]
Sekimoto, Stochastic energetics
K. Sekimoto, Stochastic energetics. Springer, 2010
2010
-
[15]
Stochastic thermodynamics: principles and perspectives,
U. Seifert, “Stochastic thermodynamics: principles and perspectives,” The Eur . Phys. J. B , vol. 64, pp. 423–431, 2008
2008
-
[16]
Stochastic thermodynamics, fluctuation theorems and molec- ular machines,
U. Seifert, “Stochastic thermodynamics, fluctuation theorems and molec- ular machines,” Rep. on Prog. in Phys. , vol. 75, no. 12, p. 126001, 2012
2012
-
[17]
Über die von der molekularkinetischen theorie der wärme geforderte bewegung von in ruhenden flüssigkeiten suspendierten teilchen,
A. Einstein, “Über die von der molekularkinetischen theorie der wärme geforderte bewegung von in ruhenden flüssigkeiten suspendierten teilchen,” Annalen der physik , vol. 4, 1905
1905
-
[18]
Persistent excitation in adaptive systems,
K. S. Narendra and A. M. Annaswamy, “Persistent excitation in adaptive systems,” Int. J. of Control , vol. 45, no. 1, pp. 127–160, 1987
1987
-
[19]
Integral concurrent learning: Adaptive control with parameter convergence using finite excitation,
A. Parikh, R. Kamalapurkar, and W. E. Dixon, “Integral concurrent learning: Adaptive control with parameter convergence using finite excitation,” Int J Adapt Control Signal Process , vol. 33, pp. 1775–1787, Dec. 2019
2019
-
[20]
Composite learning from adaptive dynamic surface control,
Y . Pan and H. Yu, “Composite learning from adaptive dynamic surface control,” IEEE Trans. Autom. Control , vol. 61, no. 9, pp. 2603–2609, 2016
2016
-
[21]
Model reference composite learning control without persistency of excitation,
Y . Pan, J. Zhang, and H. Yu, “Model reference composite learning control without persistency of excitation,” IET Control Theory & Ap- plications, vol. 10, no. 16, pp. 1963–1971, 2016
1963
-
[22]
Composite model reference adaptive control with parameter convergence under finite excitation,
N. Cho, H.-S. Shin, Y . Kim, and A. Tsourdos, “Composite model reference adaptive control with parameter convergence under finite excitation,” IEEE Trans. Autom. Control , vol. 63, pp. 811–818, 2017
2017
-
[23]
Composite learning robot control with guaranteed parrameter convergence,
Y . Pan and H. Yu, “Composite learning robot control with guaranteed parrameter convergence,” Automatica, vol. 89, pp. 415–419, Mar. 2018
2018
-
[24]
Continuous control with deep reinforcement learning,
T. P. Lillicrap, J. J. Hunt, A. Pritzel, N. Heess, T. Erez, Y . Tassa, D. Silver, and D. Wierstra, “Continuous control with deep reinforcement learning,” arXiv preprint arXiv:1509.02971 , 2015
2015 arXiv
-
[25]
Addressing function approxi- mation error in actor-critic methods,
S. Fujimoto, H. Hoof, and D. Meger, “Addressing function approxi- mation error in actor-critic methods,” in Int. Conf. on Mach. Learn. , pp. 1587–1596, 2018
2018
-
[26]
Soft actor-critic: Off- policy maximum entropy deep reinforcement learning with a stochastic actor,
T. Haarnoja, A. Zhou, P. Abbeel, and S. Levine, “Soft actor-critic: Off- policy maximum entropy deep reinforcement learning with a stochastic actor,” in Int. Conf. on Mach. Learn. , pp. 1861–1870, Pmlr, 2018
2018
-
[27]
Maximum a posteriori policy optimisation,
A. Abdolmaleki, J. T. Springenberg, Y . Tassa, R. Munos, N. Heess, and M. Riedmiller, “Maximum a posteriori policy optimisation,” arXiv preprint arXiv:1806.06920, 2018
2018 arXiv
-
[28]
The power of deeper networks for expressing natural functions,
D. Rolnick and M. Tegmark, “The power of deeper networks for expressing natural functions,” in Int. Conf. Learn. Represent. , 2018
2018
-
[29]
Why deep neural networks for function approximation?,
S. Liang and R. Srikant, “Why deep neural networks for function approximation?,” arXiv preprint arXiv:1610.04161 , 2016
2016 arXiv
-
[30]
Deep nonlinear adaptive control for unmanned aerial systems operating under dynamic uncertainties,
Z. Lamb, Z. I. Bell, M. Longmire, J. Paquet, P. Ganesh, and R. Sanfelice, “Deep nonlinear adaptive control for unmanned aerial systems operating under dynamic uncertainties,” arXiv preprint arXiv:2310.09502 , 2023
2023 arXiv
-
[31]
S. L. Brunton and J. N. Kutz, Data-driven science and engineering: Ma- chine learning, dynamical systems, and control . Cambridge University Press, 2019
2019
-
[32]
Multimodal integration learning of robot behavior using deep neural networks,
K. Noda, H. Arie, Y . Suga, and T. Ogata, “Multimodal integration learning of robot behavior using deep neural networks,” Robotics & Auton. Syst. , vol. 62, no. 6, pp. 721–736, 2014
2014
-
[33]
Detection and localization of robotic tools in robot-assisted surgery videos using deep neural networks for region proposal and detection,
D. Sarikaya, J. J. Corso, and K. A. Guru, “Detection and localization of robotic tools in robot-assisted surgery videos using deep neural networks for region proposal and detection,” IEEE Trans. Med. Imaging , vol. 36, no. 7, pp. 1542–1549, 2017
2017
-
[34]
Analytic deep neural network-based robot control,
H.-T. Nguyen and C. C. Cheah, “Analytic deep neural network-based robot control,” IEEE/ASME Trans. Mechatron., vol. 27, no. 4, pp. 2176– 2184, 2022
2022
-
[35]
Lyapunov-derived control and adaptive update laws for inner and outer layer weights of a deep neural network,
O. Patil, D. Le, M. Greene, and W. E. Dixon, “Lyapunov-derived control and adaptive update laws for inner and outer layer weights of a deep neural network,” IEEE Control Syst Lett. , vol. 6, pp. 1855–1860, 2022
2022
-
[36]
Deep residual neural network (ResNet)-based adaptive control: A Lyapunov-based approach,
O. S. Patil, D. M. Le, E. Griffis, and W. E. Dixon, “Deep residual neural network (ResNet)-based adaptive control: A Lyapunov-based approach,” in Proc. IEEE Conf. Decis. Control , pp. 3487–3492, 2022
2022
-
[37]
Deep recurrent neural network-based observer for uncertain nonlinear systems,
E. Griffis, O. Patil, W. Makumi, and W. E. Dixon, “Deep recurrent neural network-based observer for uncertain nonlinear systems,” in IF AC World Congr ., pp. 6851–6856, 2023
2023
-
[38]
Lyapunov-based dropout deep neural network (Lb-DDNN) controller,
S. Akbari, E. J. Griffis, O. S. Patil, and W. E. Dixon, “Lyapunov-based dropout deep neural network (Lb-DDNN) controller,” arXiv preprint arXiv:2310.19938, 2023
2023 arXiv
-
[40]
J. R. Magnus and H. Neudecker, Matrix differential calculus with applications in statistics and econometrics . John Wiley & Sons, 2019
2019
-
[41]
Krstic, I
M. Krstic, I. Kanellakopoulos, and P. V . Kokotovic, Nonlinear and Adaptive Control Design . New York: John Wiley & Sons, 1995
1995
-
[42]
Stochastic thermodynamics: Dissi- pativity, accumulativity, energy storage and entropy production,
M. Lanchares and W. M. Haddad, “Stochastic thermodynamics: Dissi- pativity, accumulativity, energy storage and entropy production,” Philos. Trans. of the Royal Soc. A , vol. 381, no. 2256, p. 20220284, 2023
2023
-
[43]
Billingsley, Probability and measure
P. Billingsley, Probability and measure . John Wiley & Sons, 2017
2017
-
[44]
H. J. Kushner, Stochastic stability and control. ACADEMIC PRESS, INC., 1967
1967
-
[45]
Goodfellow, Y
I. Goodfellow, Y . Bengio, and A. Courville, Deep Learning, vol. 1. MIT Press, 2016
2016
-
[46]
System identifi- cation and control using lyapunov-based deep neural networks without persistent excitation: A concurrent learning approach,
R. G. Hart, O. S. Patil, Z. I. Bell, and W. E. Dixon, “System identifi- cation and control using lyapunov-based deep neural networks without persistent excitation: A concurrent learning approach,” arXiv preprint arXiv:2505.10678, 2025
2025 arXiv
-
[47]
Fradkov, Cybernetical physics: from control of chaos to quantum control
A. Fradkov, Cybernetical physics: from control of chaos to quantum control. Springer, 2007
2007
-
[48]
Multilayer neural-net robot controller with guaranteed tracking performance,
F. L. Lewis, A. Yesildirek, and K. Liu, “Multilayer neural-net robot controller with guaranteed tracking performance,” IEEE Trans. Neural Netw., vol. 7, no. 2, pp. 388–399, 1996
1996
-
[50]
H. J. Kushner, Introduction to Stochastic Control . New York: Holt, Reinhart and Winston, 1971
1971
-
[51]
A sufficiently smooth projection operator,
Z. Cai, M. S. de Queiroz, and D. M. Dawson, “A sufficiently smooth projection operator,” IEEE Trans. Autom. Control , vol. 51, pp. 135–139, Jan. 2006
2006
-
[52]
Swish: a self-gated activation function,
P. Ramachandran, B. Zoph, and Q. V . Le, “Swish: a self-gated activation function,” arXiv preprint arXiv:1710.05941 , vol. 7, no. 1, p. 5, 2017
2017 arXiv
-
[53]
Delving deep into rectifiers: Surpassing human-level performance on imagenet classification,
K. He, X. Zhang, S. Ren, and J. Sun, “Delving deep into rectifiers: Surpassing human-level performance on imagenet classification,” in Proc. IEEE Int. Conf. Comput. Vis. , pp. 1026–1034, 2015. Saiedeh Akbari is a Ph.D. candidate in the de- partment of Mechanical and Aerospace ...
2015
Reviewed August 5, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.