Tailoring interactions between active nematic defects with reinforcement learning

Aaron R. Dinner; Carlos Floyd; Suriyanarayanan Vaikuntanathan

REVIEW 1 major objections 3 minor 52 references

Tailoring interactions between active nematic defects with reinforcement learning

T0 review · 1 major / 3 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper demonstrates that reinforcement learning can control pairs of active nematic defects with local activity fields, making them follow designer dynamics such as overdamped springs with tunable stiffness.

desk verdict A credible, honest proof-of-principle for RL control of active nematic defects, undercut mainly by missing code, typos, and an untested generality claim. read the letter →

arxiv 2411.09588 v1 pith:FOLFVZGX submitted 2024-11-14 cond-mat.soft

Carlos Floyd , Aaron R. Dinner , Suriyanarayanan Vaikuntanathan This is my paper · ORCID

classification cond-mat.soft

keywords activenematicstopologicaldefectsreinforcementlearningactivityfieldsfeedbackcontroldefectdynamicsoverdampedspringsmodel-free

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper shows that reinforcement learning can program the motion of topological defects in active nematic fluids by shaping the local activity field, without needing an explicit model of the hydrodynamics. The authors train a controller to adjust a small set of activity parameters, such as amplitude, offset, or angular placement, so that a defect pair follows prescribed dynamical laws like the exponential relaxation of an overdamped spring with a chosen stiffness. Because the controller sees only a coarse projection of the field, a scalar separation or orientation, and acts through low-dimensional activity profiles, the authors argue that such feedback loops are simple enough to be realized experimentally with light-controlled molecular motors and, plausibly, inside cells. If the approach holds up, it offers a general route to designing interactions between defects rather than measuring and modeling them.

What carries the argument

The load-bearing machinery is a closed feedback loop in which a neural-network policy, trained with the deep deterministic policy gradient algorithm, maps a coarse state, the defect separation rsep or the orientation variable ζ+ = sin(φ+), to an action that parameterizes the activity field α(r) near one defect. The activity field couples to defect motion through known physics: +1/2 defects propel along their orientation with a velocity approximately proportional to local activity, and -1/2 defects respond to second-order and higher gradients of activity. The policy is trained purely from episode rewards that measure how closely the realized dynamics match a target law such as ˙r∗sep = -k0(rsep - l0). The same machinery is reused across the three tasks by changing the observable, the action parameterization, and the reward.

What would settle it

Retrain the spring task with the state augmented to include the local director field around the defects, or with varied initial -1/2 orientations; if the fitted spring constant k0 changes materially or the policy fails to track the target law, the scalar projection was insufficient.

Watch

Extended reading notes

Core claim

The central claim is that local, spatiotemporally varying activity fields can induce desired interactions between a +1/2 and a -1/2 nematic defect, and that reinforcement learning can discover the activity protocol achieving this. The paper demonstrates three proof-of-principle tasks: holding a defect pair at a target separation by adjusting the activity amplitude on a +1/2 defect; making the separation relax to a rest length with a tunable overdamped spring constant by adjusting the amplitude and offset of an activity field near a -1/2 defect; and making the +1/2 defect's orientation relax to zero at a tunable rate by rotating the angular placement of the activity field. In each case the fitted decay rate or target separation matches the requested value across a range of parameters. The authors conclude that model-free reinforcement learning can impose designer defect dynamics and that the low-dimensional state and action spaces are sufficient, supporting the picture that a few collective degrees of freedom capture the relevant defect response.

Load-bearing premise

The paper's approach assumes that the scalar state and the restricted family of activity profiles used for each task contain enough information to determine the defect response; if unobserved details such as the -1/2 defect's orientation or local flow variations matter, the trained policies would only work under the narrow conditions they were trained on.

Editorial extensions

If this is right

Local activity patterning is a sufficient control channel to make active nematic defects obey user-specified interaction laws, not just static configurations.
Because the controller is model-free and uses only coarse observations, the same training procedure could transfer to experimental active nematics driven by light-activated motors without re-deriving hydrodynamic parameters.
Demonstrated control of both positional and orientational defect dynamics opens the way to imposing designer pair interactions, such as springs with tunable stiffness, as effective forces between defects.
The success of low-dimensional state and action spaces supports the effective one- and two-body defect equations used in theory and motivates simple feedback rules for tissue-scale control.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the scalar-state sufficiency generalizes, biological feedback loops for defect positioning could be minimal: one measured quantity feeding one activity channel, making optogenetic implementation in tissues more plausible than full-field controllers.
The same reinforcement-learning loop could impose non-exponential and non-monotone interaction laws, such as a Lennard-Jones-like potential or a repulsive barrier, since the reward only needs to specify the target trajectory.
A direct testable extension is to port trained policies from the overdamped simulation to an experimental active nematic with light-controlled myosins and compare defect separation trajectories against the prescribed exponential decay.
The finding that imperfect, low-dimensional feedback suffices suggests that other active materials with slow collective variables could be steered by similar coarse feedback loops.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

This is a straightforward, honest proof-of-principle paper. The genuinely new bit is that the authors train a deep RL policy to control a pair of ±1/2 defects in a simulated active nematic, using very low-dimensional state and action spaces, and get the defects to follow designer overdamped-spring dynamics in both their separation and the +1/2 defect's orientation. That is a real addition to the active-matter control literature, which has mostly used optimal control or hand-designed protocols, and RL on other active systems has targeted static properties rather than prescribed dynamics.

The paper does several things well. The three tasks are cleanly separated, the simulation details are reported in the appendix, and the results are averaged over 10 seeds with standard deviations. The authors disclose the high-stiffness deviation honestly. The verification procedure (fitting an exponential to the approach to the target) is a reasonable way to check that the learned dynamics match the target, and the circularity concern is minor: the reward is defined against the target at each step, so the exponential fit is in some sense checking the training objective, but the fact that RL actually succeeds in the nonlinear system is the nontrivial part.

The soft spots are mostly presentation and provenance. Section III C has a clear numeric typo: kfit_theta = 0.0067 vs target 0.0007 (later in the caption it correctly says 0.00067), and Figure 6D caption repeats 'rsep' when the task is about sin(phi). No code or data is provided, which is a real hindrance for a paper whose method is computational. The deeper caveat is the one the stress-test flags: the low-dimensional state/action sufficiency is shown only for a narrow family of initial conditions (vertical alignment, a limited separation range, and small director rotations). The authors call it a proof of principle and say 'suggests' in the abstract, so they are not overclaiming, but the reader should not take the 'coarse projections suffice for precise feedback control' line as established beyond these tasks.

Overall, I think the paper is sound at the level of a simulation proof-of-concept. It deserves a serious referee, not a desk reject. I would ask for the code and data, a fix of the typos, and ideally one generalization test (e.g., a different initial orientation or a task with two state variables) to sharpen the claim. I'd also suggest the authors soften the abstract's final sentence to make clear that this is a demonstration, not a general result.

Referee Report

1 major / 3 minor

Summary. The paper uses deep deterministic policy gradient (DDPG) reinforcement learning to train closed-loop policies that control pairs of ±1/2 nematic defects in a simulated overdamped active nematic. Three tasks are considered: controlling the horizontal separation of a vertically aligned pair to a target value by adjusting the amplitude of an activity disk on the +1/2 defect; making the separation relax to a rest length with a prescribed exponential rate by adjusting the amplitude and offset of an activity field near the −1/2 defect; and making sin(φ+) of the +1/2 defect relax to zero with a prescribed rate by rotating the angular position of an activity field near the defect. For each task the authors train with multiple random seeds, fit the realized relaxation rate, and compare the fitted rate to the target rate, reporting standard deviations. They conclude that model-free RL with coarse, low-dimensional state/action representations can impose designer defect interaction laws.

Significance. The demonstration is credible and useful as a proof of principle. The paper includes multiple seeds, standard deviations, quantitative comparisons of fitted rates to target rates, and acknowledges the known breakdown at high stiffness. It also carefully separates the reward construction from the evaluation metric, so the fitted exponential rate is not used to train the policy; the reported agreement is a nontrivial check. Because the method is model-free at the policy level and uses only coarse observables, the results suggest a practical route to optogenetic control of active nematics and support the plausibility of simple feedback loops in biological contexts. The main limitation is that the sufficiency of the low-dimensional state/action representation is demonstrated only for a narrow family of initial conditions, and the paper would benefit from explicit out-of-distribution tests or more cautious wording.

major comments (1)

[II B and IV] The conclusion that a scalar state such as S=(rsep−l0)/50 or ζ+ is a sufficient statistic for the defect response is not tested outside the training distribution. In task 2, episodes always begin with vertically aligned defects whose initial horizontal separation is drawn from [37.5,62.5]; in task 3, the director is rotated uniformly by at most ±0.4 rad from a vertically aligned, separation-50 configuration. Consequently, unobserved degrees of freedom (defect orientations, vertical offsets, flow history) are confounded with the scalar state, and the trained policy could be a lookup table over a low-dimensional manifold rather than a general feedback law. I ask the authors to test the trained policies on configurations outside this family—for example, task 2 with rotated pair axes or nonzero vertical offsets, and task 3 with initial separations outside 50 or director rotations larger than 0.4 rad—and report quantitative success or failure. If performance degrades, the abstract and Discussion should be revised to state that coarse projections suffice within the specific configuration family tested, rather than as a general property.

minor comments (3)

[III C and Fig. 6] The text reports k_fit_theta = 0.0067 for a target k_theta = 0.0007, while Fig. 6B labels the fit as 0.00067; the factor-of-ten discrepancy should be corrected. In addition, the Fig. 6D caption refers to 'trajectories of rsep', but the panels plot sin(φ+) (ζ+), which is the quantity described in the text.
[II C and III] The paper does not include a code or data availability statement. Since the active nematic solver is custom and the RL results depend on the specific simulator, releasing the code (or at least a documented repository) would materially improve reproducibility.
[III B] The deviations at high k0 are mentioned in the text and visible in Fig. 5E, but the paper does not quantify how large these deviations become or whether they occur consistently across seeds; a brief numerical statement would help the reader judge the practical range of achievable stiffnesses.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the designer laws enter only via the explicitly user-specified reward, and the exponential-fit verifications are post-hoc checks of learned policies against the full PDE simulation, never fitted inputs to training.

full rationale

The paper's central demonstration — that RL can learn feedback policies making a pair of ±1/2 defects in a simulated active nematic follow prescribed positional and orientational dynamical laws — is self-contained and does not reduce to its inputs. The designer laws (Eqs. 5 and 8) are explicitly user-specified targets, not discovered outputs; they enter training only through the reward functions (Eq. 7 and the analogous ζ+ reward in Sec. III C). The policies themselves are produced by DDPG trial-and-error interaction with the full active-nematic PDE (Eqs. 10-15), which the agent never observes in closed form. The reported verification fits an exponential to the resulting trajectories and compares the fitted rates (kfit_0, kfit_θ) with the target rates; these fitted rates are used only for post-hoc comparison and are never fed back into training, so no fitted parameter is renamed as a prediction. The agreement is therefore a genuine measure of whether RL succeeded at the control task, and it is not guaranteed by construction: the paper explicitly documents failure at high stiffness ('the policy has difficulty pulling the defects faster than the nematic material timescales allow'), which shows the match at accessible rates is a nontrivial statement about the controllability of the simulated system under the allowed two-scalar action family. Task 1 is likewise an honest control report: the reward encodes the target separation l0 and the verification (lfit_0 ≈ l0) confirms the learned policy reached it. The manuscript's self-citations (Ref. 32 for prior imperfect-feedback work by the same authors; Refs. 38-40 for the numerical solver and model) are contextual or tool-level and are not load-bearing for the RL demonstration, which is evaluated against independent simulation. The broader claim that low-dimensional projections suffice for feedback control extrapolates from tasks in which the controlled variable is also the state variable; that is a scope and external-validity concern, not circularity.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The central claim is a simulation demonstration, so the ledger contains hand-chosen control parameterizations and domain assumptions inherited from prior active nematic models. No new physical entities are introduced. The key assumption is that coarse projections of the nematic field suffice for feedback control.

free parameters (4)

Activity profile shape (c, m) in Eq. 4 = c=5, m=1 for Tasks 1 and 2; c=5, m=1 with amplitude -7.5 for Task 3
Chosen to provide smooth localized activity; not learned; task outcomes may depend on these hand-set values.
State normalization scale = 50
Used in S=(rsep-l0)/50 to scale the state input; affects learning dynamics but not the physical claim.
Action ranges = [-5,5] (Task 1), [0,12] amplitude and [-10,10] offset (Task 2)
Hand-set bounds on activity amplitude and offset; the policy operates within these limits.
RL hyperparameters = learning rate 0.001, batch 32, gamma 0.99, rho 0.995, noise std 0.05, 2 hidden layers of 32 units
Standard but hand-tuned; results may be sensitive to these choices.

assumptions (3)

domain assumption The active nematic obeys Eqs. (10)-(11) with the Landau-de Gennes free energy Eq. (15) and parameters xi=0.7, A0=0.1, U=3.5, Gamma_H=1.5, L=0.1, gamma_v=10.
All training and evaluation simulations run inside this model, imported from Refs. 16, 40, 49; the paper does not validate it against new experiments.
domain assumption Defect positions and orientations are computed via Ref. 33, and the coarse observables (rsep, zeta+) are sufficient to determine the effect of the chosen activity fields.
This is the load-bearing representation assumption; it is supported only by the demonstrated task success, not proven ahead of time.
standard math Deep deterministic policy gradient will converge to a good deterministic policy with the chosen hyperparameters and reward.
Standard RL practice; no convergence guarantee for this nonlinear PDE control problem is established, but empirical learning curves are shown.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Tailoring interactions between active nematic defects with reinforcement learning." pith.science (2026). https://pith.science/paper/FOLFVZGX

@misc{pith2026241109588,
  author       = {Pith},
  title        = {Pith review of: Tailoring interactions between active nematic defects with reinforcement learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FOLFVZGX}},
  note         = {Machine review of arXiv:2411.09588}
}

read the original abstract

Active nematics, formed from a liquid crystalline suspension of active force dipoles, are a paradigmatic active matter system whose study provides insights into how chemical driving produces the cellular mechanical forces essential for life. Recent advances in optogenetic control over molecular motors and cell-signaling pathways now allow experimenters to mimic the spatiotemporal regulation of activity necessary to drive biologically relevant active nematic flows in vivo. However, engineering effective activity protocols remains challenging due to the system's complex dynamics. Here, we explore a model-free approach for controlling active nematic fields using reinforcement learning. Specifically, we demonstrate how local activity fields can induce interactions between pairs of nematic defects, enabling them to follow designer dynamical laws such as those of overdamped springs with varying stiffnesses. Reinforcement learning bypasses the need for accurate parameterization and model representation of the nematic system, and could thus transfer straightforwardly to experimental implementation. Moreover, the sufficiency of our low-dimensional system observables and actions suggests that coarse projections of the active nematic field can be used for precise feedback control, making the biological implementation of such feedback loops plausible.

Figures

Figures reproduced from arXiv: 2411.09588 by the authors.

**Figure 1.** FIG. 1. Schematic overview of the closed-loop control over active [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 3.** FIG. 3. Schematic illustration of the structure of an experiment, [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

Figures from the paper (3 more)

**Figure 4.** Figure 4: FIG. 4. Results for task 1. A) Schematic illustration of the task, in which a nearly uniform disk of activity is applied to the center of the [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

**Figure 5.** Figure 5: FIG. 5. Results for task 2. A) Schematic illustration of the task, in which an inhomogeneous activity is field applied near the [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]

**Figure 6.** Figure 6: FIG. 6. Results for task 3. A) Schematic illustration of the task, in which an inhomogeneous activity is field applied near the [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

52 extracted references · 48 canonical work pages

[1]

Cell mechanics and the cytoskeleton

Daniel A Fletcher and R Dyche Mullins. Cell mechanics and the cytoskeleton. Nature , 463(7280):485--492, 2010

work page 2010
[2]

The actin cytoskeleton as an active adaptive material

Shiladitya Banerjee, Margaret L Gardel, and Ulrich S Schwarz. The actin cytoskeleton as an active adaptive material. Annual review of condensed matter physics , 11(1):421--439, 2020

work page 2020
[3]

Physics of smart active matter: integrating active matter and control to gain insights into living systems

Herbert Levine and Daniel I Goldman. Physics of smart active matter: integrating active matter and control to gain insights into living systems. Soft Matter , 19(23):4204--4207, 2023

work page 2023
[4]

Cytokinesis: placing and making the final cut

Francis A Barr and Ulrike Gruneberg. Cytokinesis: placing and making the final cut. Cell , 131(5):847--860, 2007

work page 2007
[5]

Topological defects in the nematic order of actin fibres as organization centres of hydra morphogenesis

Yonit Maroudas-Sacks, Liora Garion, Lital Shani-Zerbib, Anton Livshits, Erez Braun, and Kinneret Keren. Topological defects in the nematic order of actin fibres as organization centres of hydra morphogenesis. Nature Physics , 17(2):251--259, 2021

work page 2021
[6]

Integer topological defects organize stresses driving tissue morphogenesis

Pau Guillamat, Carles Blanch-Mercader, Guillaume Pernollet, Karsten Kruse, and Aur \'e lien Roux. Integer topological defects organize stresses driving tissue morphogenesis. Nature materials , 21(5):588--597, 2022

work page 2022
[7]

Defect-mediated dynamics of coherent structures in active nematics

Mattia Serra, Linnea Lemma, Luca Giomi, Zvonimir Dogic, and Lakshminarayanan Mahadevan. Defect-mediated dynamics of coherent structures in active nematics. Nature Physics , 19(9):1355--1361, 2023

work page 2023
[8]

Topological active matter

Suraj Shankar, Anton Souslov, Mark J Bowick, M Cristina Marchetti, and Vincenzo Vitelli. Topological active matter. Nature Reviews Physics , 4(6):380--398, 2022

work page 2022

Show all 52 references

[9]

Tissue active matter: Integrating mechanics and signaling into dynamical models

David B Br \"u ckner and Edouard Hannezo. Tissue active matter: Integrating mechanics and signaling into dynamical models. Cold Spring Harbor Perspectives in Biology , page a041653, 2024

2024
[10]

Visceral organ morphogenesis via calcium-patterned muscle constrictions

Noah P Mitchell, Dillon J Cislo, Suraj Shankar, Yuzheng Lin, Boris I Shraiman, and Sebastian J Streichan. Visceral organ morphogenesis via calcium-patterned muscle constrictions. Elife , 11:e77355, 2022

2022
[11]

Mechanical and biochemical feedback combine to generate complex contractile oscillations in cytokinesis

Michael E Werner, Dylan D Ray, Coleman Breen, Michael F Staddon, Florian Jug, Shiladitya Banerjee, and Amy Shaub Maddox. Mechanical and biochemical feedback combine to generate complex contractile oscillations in cytokinesis. Current Biology , 34(14):3201--3214, 2024

2024
[12]

A minimal mathematical model for polarity establishment and centralspindlin-independent cytokinesis

Ondrej Maxian, Katrina M Longhini, and Michael Glotzer. A minimal mathematical model for polarity establishment and centralspindlin-independent cytokinesis. bioRxiv , pages 2024--08, 2024

2024
[13]

Disordered actomyosin networks are sufficient to produce cooperative and telescopic contractility

Ian Linsmeier, Shiladitya Banerjee, Patrick W Oakes, Wonyeong Jung, Taeyoon Kim, and Michael P Murrell. Disordered actomyosin networks are sufficient to produce cooperative and telescopic contractility. Nature communications , 7(1):12615, 2016

2016
[14]

Engineering myosins for long-range transport on actin filaments

Tony D Schindler, Lu Chen, Paul Lebel, Muneaki Nakamura, and Zev Bryant. Engineering myosins for long-range transport on actin filaments. Nature nanotechnology , 9(1):33--38, 2014

2014
[15]

Remote control of myosin and kinesin motors using light-activated gearshifting

Muneaki Nakamura, Lu Chen, Stuart C Howes, Tony D Schindler, Eva Nogales, and Zev Bryant. Remote control of myosin and kinesin motors using light-activated gearshifting. Nature nanotechnology , 9(9):693--697, 2014

2014
[16]

Spatiotemporal control of liquid crystal structure and dynamics through activity patterning

Rui Zhang, Steven A Redford, Paul V Ruijgrok, Nitin Kumar, Ali Mozaffari, Sasha Zemsky, Aaron R Dinner, Vincenzo Vitelli, Zev Bryant, Margaret L Gardel, et al. Spatiotemporal control of liquid crystal structure and dynamics through activity patterning. Nature materials , 20(6)...

2021
[17]

Spatio-temporal patterning of extensile active stresses in microtubule-based active fluids

Linnea M Lemma, Minu Varghese, Tyler D Ross, Matt Thomson, Aparna Baskaran, and Zvonimir Dogic. Spatio-temporal patterning of extensile active stresses in microtubule-based active fluids. PNAS nexus , 2(5):pgad130, 2023

2023
[18]

Shining a light on rhoa: Optical control of cell contractility

Shreya Chandrasekar, Jordan R Beach, and Patrick W Oakes. Shining a light on rhoa: Optical control of cell contractility. The International Journal of Biochemistry & Cell Biology , 161:106442, 2023

2023
[19]

Optimal control of active nematics

Michael M Norton, Piyush Grover, Michael F Hagan, and Seth Fraden. Optimal control of active nematics. Physical review letters , 125(17):178005, 2020

2020
[20]

Spatiotemporal control of structure and dynamics in a polar active fluid

Saptorshi Ghosh, Chaitanya Joshi, Aparna Baskaran, and Michael F Hagan. Spatiotemporal control of structure and dynamics in a polar active fluid. arXiv preprint arXiv:2405.07942 , 2024

2024 arXiv
[21]

Achieving designed texture and flows in bulk active nematics using optimal control theory

Saptorshi Ghosh, Aparna Baskaran, and Michael F Hagan. Achieving designed texture and flows in bulk active nematics using optimal control theory. arXiv preprint arXiv:2408.14596 , 2024

2024 arXiv
[22]

Exact coherent structures and phase space geometry of preturbulent 2d active nematic channel flow

Caleb G Wagner, Michael M Norton, Jae Sung Park, and Piyush Grover. Exact coherent structures and phase space geometry of preturbulent 2d active nematic channel flow. Physical Review Letters , 128(2):028003, 2022

2022
[23]

Design rules for controlling active topological defects

Suraj Shankar, Luca VD Scharrer, Mark J Bowick, and M Cristina Marchetti. Design rules for controlling active topological defects. Proceedings of the National Academy of Sciences , 121(21):e2400933121, 2024

2024
[24]

Dislocation reactions, grain boundaries, and irreversibility in two-dimensional lattices using topological tweezers

William TM Irvine, Andrew D Hollingsworth, David G Grier, and Paul M Chaikin. Dislocation reactions, grain boundaries, and irreversibility in two-dimensional lattices using topological tweezers. Proceedings of the National Academy of Sciences , 110(39):15544--15548, 2013

2013
[25]

Reinforcement learning: an introduction

Richard S Sutton. Reinforcement learning: an introduction. A Bradford Book , 2018

2018
[26]

Control theory for physicists

John Bechhoefer. Control theory for physicists . Cambridge University Press, 2021

2021
[27]

Learning to control active matter

Martin J Falk, Vahid Alizadehyazdi, Heinrich Jaeger, and Arvind Murugan. Learning to control active matter. Physical Review Research , 3(3):033291, 2021

2021
[28]

Probing the theoretical and computational limits of dissipative design

Shriram Chennakesavalu and Grant M Rotskoff. Probing the theoretical and computational limits of dissipative design. The Journal of Chemical Physics , 155(19), 2021

2021
[29]

Adaptive nonequilibrium design of actin-based metamaterials: Fundamental and practical limits of control

Shriram Chennakesavalu, Sreekanth K Manikandan, Frank Hu, and Grant M Rotskoff. Adaptive nonequilibrium design of actin-based metamaterials: Fundamental and practical limits of control. Proceedings of the National Academy of Sciences , 121(8):e2310238121, 2024

2024
[30]

Active particles bound by information flows

Utsab Khadka, Viktor Holubec, Haw Yang, and Frank Cichos. Active particles bound by information flows. Nature communications , 9(1):3864, 2018

2018
[31]

Self-organization of active particles by quorum sensing rules

Tobias B \"a uerle, Andreas Fischer, Thomas Speck, and Clemens Bechinger. Self-organization of active particles by quorum sensing rules. Nature communications , 9(1):3232, 2018

2018
[32]

Learning to control non-equilibrium dynamics using local imperfect gradients

Carlos Floyd, Aaron R Dinner, and Suriyanarayanan Vaikuntanathan. Learning to control non-equilibrium dynamics using local imperfect gradients. arXiv preprint arXiv:2404.03798 , 2024

2024 arXiv
[33]

Orientation of topological defects in 2d nematic liquid crystals

Xingzhou Tang and Jonathan V Selinger. Orientation of topological defects in 2d nematic liquid crystals. Soft matter , 13(32):5481--5490, 2017

2017
[34]

Defect unbinding in active nematics

Suraj Shankar, Sriram Ramaswamy, M Cristina Marchetti, and Mark J Bowick. Defect unbinding in active nematics. Physical review letters , 121(10):108002, 2018

2018
[35]

Data-driven discovery of active nematic hydrodynamics

Chaitanya Joshi, Sattvic Ray, Linnea M Lemma, Minu Varghese, Graham Sharp, Zvonimir Dogic, Aparna Baskaran, and Michael F Hagan. Data-driven discovery of active nematic hydrodynamics. Physical review letters , 129(25):258001, 2022

2022
[36]

Deterministic policy gradient algorithms

David Silver, Guy Lever, Nicolas Heess, Thomas Degris, Daan Wierstra, and Martin Riedmiller. Deterministic policy gradient algorithms. In International conference on machine learning , pages 387--395. Pmlr, 2014

2014
[37]

Continuous control with deep reinforcement learning

Timothy P Lillicrap, Jonathan J Hunt, Alexander Pritzel, Nicolas Heess, Tom Erez, Yuval Tassa, David Silver, and Daan Wierstra. Continuous control with deep reinforcement learning. arXiv preprint arXiv:1509.02971 , 2015

2015 arXiv
[38]

Simulating structured fluids with tensorial viscoelasticity

Carlos Floyd, Suriyanarayanan Vaikuntanathan, and Aaron R Dinner. Simulating structured fluids with tensorial viscoelasticity. The Journal of Chemical Physics , 158(5), 2023

2023
[39]

Pattern formation in odd viscoelastic fluids

Carlos Floyd, Aaron R Dinner, and Suriyanarayanan Vaikuntanathan. Pattern formation in odd viscoelastic fluids. Physical Review Research , 6(3):033100, 2024

2024
[40]

Motor crosslinking augments elasticity in active nematics

Steven A Redford, Jonathan Colen, Jordan L Shivers, Sasha Zemsky, Mehdi Molaei, Carlos Floyd, Paul V Ruijgrok, Vincenzo Vitelli, Zev Bryant, Aaron R Dinner, et al. Motor crosslinking augments elasticity in active nematics. Soft Matter , 20(11):2480--2490, 2024

2024
[41]

Correlation lengths in hydrodynamic models of active nematics

Ewan J Hemingway, Prashant Mishra, M Cristina Marchetti, and Suzanne M Fielding. Correlation lengths in hydrodynamic models of active nematics. Soft Matter , 12(38):7943--7952, 2016

2016
[42]

Reinforcementlearning.jl: A reinforcement learning package for the julia programming language, 2020

Jun Tian and other contributors. Reinforcementlearning.jl: A reinforcement learning package for the julia programming language, 2020

2020
[43]

Defect dynamics in active nematics

Luca Giomi, Mark J Bowick, Prashant Mishra, Rastko Sknepnek, and M Cristina Marchetti. Defect dynamics in active nematics. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences , 372(2029):20130365, 2014

2014
[44]

Reaching the limit in autonomous racing: Optimal control versus reinforcement learning

Yunlong Song, Angel Romero, Matthias M \"u ller, Vladlen Koltun, and Davide Scaramuzza. Reaching the limit in autonomous racing: Optimal control versus reinforcement learning. Science Robotics , 8(82):eadg1462, 2023

2023
[45]

A machine learning approach to robustly determine director fields and analyze defects in active nematics

Yunrui Li, Zahra Zarei, Phu N Tran, Yifei Wang, Aparna Baskaran, Seth Fraden, Michael F Hagan, and Pengyu Hong. A machine learning approach to robustly determine director fields and analyze defects in active nematics. Soft matter , 20(8):1869--1883, 2024

2024
[46]

Deep-learning optical flow outperforms piv in obtaining velocity fields from active nematics

Phu N Tran, Sattvic Ray, Linnea Lemma, Yunrui Li, Reef Sweeney, Aparna Baskaran, Zvonimir Dogic, Pengyu Hong, and Michael F Hagan. Deep-learning optical flow outperforms piv in obtaining velocity fields from active nematics. arXiv preprint arXiv:2404.15497 , 2024

2024 arXiv
[47]

Mastering the game of go with deep neural networks and tree search

David Silver, Aja Huang, Chris J Maddison, Arthur Guez, Laurent Sifre, George Van Den Driessche, Julian Schrittwieser, Ioannis Antonoglou, Veda Panneershelvam, Marc Lanctot, et al. Mastering the game of go with deep neural networks and tree search. nature , 529(7587):484--489, 2016

2016
[48]

Glider soaring via reinforcement learning in the field

Gautam Reddy, Jerome Wong-Ng, Antonio Celani, Terrence J Sejnowski, and Massimo Vergassola. Glider soaring via reinforcement learning in the field. Nature , 562(7726):236--239, 2018

2018
[49]

Machine learning active-nematic hydrodynamics

Jonathan Colen, Ming Han, Rui Zhang, Steven A Redford, Linnea M Lemma, Link Morgan, Paul V Ruijgrok, Raymond Adkins, Zev Bryant, Zvonimir Dogic, et al. Machine learning active-nematic hydrodynamics. Proceedings of the National Academy of Sciences , 118(10):e2016708118, 2021

2021
[50]

Rheology of active-particle suspensions

Yashodhan Hatwalne, Sriram Ramaswamy, Madan Rao, and R Aditi Simha. Rheology of active-particle suspensions. Physical review letters , 92(11):118101, 2004

2004
[51]

Hydrodynamics of soft active matter

M Cristina Marchetti, Jean-Fran c ois Joanny, Sriram Ramaswamy, Tanniemola B Liverpool, Jacques Prost, Madan Rao, and R Aditi Simha. Hydrodynamics of soft active matter. Reviews of modern physics , 85(3):1143--1189, 2013

2013
[52]

Thermodynamics of flowing systems: with internal microstructure

Antony N Beris and Brian J Edwards. Thermodynamics of flowing systems: with internal microstructure . Number 36. Oxford University Press, USA, 1994

1994

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Cell mechanics and the cytoskeleton

Daniel A Fletcher and R Dyche Mullins. Cell mechanics and the cytoskeleton. Nature , 463(7280):485--492, 2010

work page 2010

[2] [2]

The actin cytoskeleton as an active adaptive material

Shiladitya Banerjee, Margaret L Gardel, and Ulrich S Schwarz. The actin cytoskeleton as an active adaptive material. Annual review of condensed matter physics , 11(1):421--439, 2020

work page 2020

[3] [3]

Physics of smart active matter: integrating active matter and control to gain insights into living systems

Herbert Levine and Daniel I Goldman. Physics of smart active matter: integrating active matter and control to gain insights into living systems. Soft Matter , 19(23):4204--4207, 2023

work page 2023

[4] [4]

Cytokinesis: placing and making the final cut

Francis A Barr and Ulrike Gruneberg. Cytokinesis: placing and making the final cut. Cell , 131(5):847--860, 2007

work page 2007

[5] [5]

Topological defects in the nematic order of actin fibres as organization centres of hydra morphogenesis

Yonit Maroudas-Sacks, Liora Garion, Lital Shani-Zerbib, Anton Livshits, Erez Braun, and Kinneret Keren. Topological defects in the nematic order of actin fibres as organization centres of hydra morphogenesis. Nature Physics , 17(2):251--259, 2021

work page 2021

[6] [6]

Integer topological defects organize stresses driving tissue morphogenesis

Pau Guillamat, Carles Blanch-Mercader, Guillaume Pernollet, Karsten Kruse, and Aur \'e lien Roux. Integer topological defects organize stresses driving tissue morphogenesis. Nature materials , 21(5):588--597, 2022

work page 2022

[7] [7]

Defect-mediated dynamics of coherent structures in active nematics

Mattia Serra, Linnea Lemma, Luca Giomi, Zvonimir Dogic, and Lakshminarayanan Mahadevan. Defect-mediated dynamics of coherent structures in active nematics. Nature Physics , 19(9):1355--1361, 2023

work page 2023

[8] [8]

Topological active matter

Suraj Shankar, Anton Souslov, Mark J Bowick, M Cristina Marchetti, and Vincenzo Vitelli. Topological active matter. Nature Reviews Physics , 4(6):380--398, 2022

work page 2022