Feedback percolation on complex networks
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.
Kai: I'm Kai.
Mira: And I'm Mira, here with Lev.
Kai: We're looking at the paper: "Feedback percolation on complex networks".
Mira: Here's the gist of the paper. Feedback percolation introduces an unified framework that dynamically couples the microscopic activation probability to the macroscopic size of the giant component pg1.
Kai: This simple feedback mechanism produces a rich variety of behaviors both analytically and numerically, establishing that macroscopic feedback provides a unifying physical mechanism for phenomena ranging from self-regulating oscillations to systemic infrastructure collapse pg1.
Mira: Traditional percolation theory assumes static microscopic rules, limiting its ability to describe real-world complex systems where macroscopic order actively regulates local interactions pg1.
Kai: The authors introduce feedback percolation, which dynamically couples the microscopic activation probability to the macroscopic size of the giant component pg1.
Paper summary: Kai: The core idea here is that traditional percolation theory is too simple because it ignores how macroscopic order actively regulates local interactions >
Mira: They introduce feedback percolation as a way to fix that by dynamically linking the microscopic activation probability to the size of the giant component, which they show can lead to a rich variety of behaviors analytically and numerically >
Lev: So they're suggesting that this feedback isn't just some fancy trick, it’s a fundamental driver for diverse percolation transitions like discontinuous jumps and stable oscillatory states >
Kai: They define this feedback through an iterative update rule, where the probability of activation changes based on the size of the giant component from the previous step >
Mira: And they show that depending on what that feedback function is—what they call f(S)—you can get things like explosive jumps, hybrid transitions, limit-cycle oscillations, and even routes to chaos >
Lev: It’s interesting to hear that they've cast these iterative equations into a one dimensional map called Sn equals h of Sn minus one where the function h doesn't depend on n >
Kai: That suggests the underlying dynamics are governed by a fixed map regardless of how many steps you take in the process, which simplifies the analysis somewhat >
Mira: And they derive some specific self-consistency equations for networks with any degree distribution P of k, and they find that the percolation threshold pc for this system is determined by pc plus f(zero) equals one over G prime one of one > <ref:2603.22089#pg1>
Conclusion: Kai: So, looking at this paper, "Feedback percolation on complex networks," what they really want to say is that when you look at how big a system gets, that size actually feeds back and changes the rules for how the parts of the system interact >
Mira: They’re moving beyond just static models where everything is fixed and showing that this dynamic coupling between local activation probability and global connectivity gives us a much richer picture of what complex systems can do >
Lev: For those of us building error correction for real hardware, it means we have to consider these kinds of feedback loops because they can introduce unpredictable dynamics that standard static models miss >
Kai: It suggests that the way a giant component evolves is not just a simple growth curve but something much more complicated and self-regulating >
Mira: The implication is that understanding this unified framework could help us interpret cascading failures in interdependent networks because this feedback mechanism offers an alternative interpretation to those failures >
Hoseung Jang, *Ginestra Bianconi, *Byungjoon Min
Advanced-Basic-Convergence Research Institute · Centre for Complex Systems, School of Mathematical Sciences, Queen Mary University of London · Department of Physics, Chungbuk National University
cond-mat.stat-mech, cond-mat.dis-nn, nlin.AO, physics.soc-ph
Submitted: 2026-03-23
Updated: 2026-11-06
Comments: 10 pages, 5 figures
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 74/100
The gist: Feedback percolation introduces an unified framework that dynamically couples the microscopic activation probability to the macroscopic size of the giant component pg1.
Key concepts
- Feedback Percolation
- A framework that dynamically links the microscopic chance of activation in a network to the size of its largest connected part. This means the network's rules constantly adjust based on how big it currently is, allowing for self-regulation.
- Giant Component Size (Sn)
- This refers to the size of the largest connected cluster within a network, which is determined by standard percolation rules using an updated probability. This size acts as the macroscopic variable that feeds back into the system's local activation probability.
- Feedback Function f(S)
- This function dictates how the network's probability update depends on its current giant component size (S). Different choices for this function lead to very different outcomes, such as stable states, oscillations, or chaotic behavior in the system's evolution.
Terminology
Summary
Feedback percolation introduces an unified framework that dynamically couples the microscopic activation probability to the macroscopic size of the giant component pg1. p. 0 Feedback percolation on complex networks
Hoseung Jang,1 Ginestra Bianconi,2, ∗ and Byungjoon Min1, 3, †
1Advanced-Basic-Convergence Research Institute, Chungbuk National University, Cheongju, Chungbuk 28644, Korea
2Centre for Complex Systems, School of Mathematical Sciences,
Queen Mary University of London, Mile End Road, London, E1 4NS, UK
3Department of Physics, Chungbuk National University, Cheongju, Chungbuk 28644, Korea
(Dated: March 24, 2026)
Traditional percolation theory assumes static microscopic rules, limiting its ability to describe
real-world complex systems where macroscopic order actively regulates local interactions. Here,
we introduce feedback percolation, an unified framework that dynamically couples the microscopic
activation probability to the macroscopic size of the giant component. Depending
on the feedback functions, the system exhibits explosive discontinuous jumps, hybrid transitions,
limit-cycle oscillations, and routes to chaos, absent in classical percolation. Our findings establish
that macroscopic feedback provides a unifying physical mechanism for phenomena ranging from
self-regulating oscillations to systemic infrastructure collapse.Where the paper says it
Where the paper says it
Introduction to the Framework
Traditional percolation theory assumes static microscopic rules, limiting its ability to describe real-world complex systems where macroscopic order actively regulates local interactions pg1. The authors introduce feedback percolation, which dynamically couples the microscopic activation probability to the macroscopic size of the giant component pg1. p. 0 Feedback percolation on complex networks
Hoseung Jang,1 Ginestra Bianconi,2, ∗ and Byungjoon Min1, 3, †
1Advanced-Basic-Convergence Research Institute, Chungbuk National University, Cheongju, Chungbuk 28644, Korea
2Centre for Complex Systems, School of Mathematical Sciences,
Queen Mary University of London, Mile End Road, London, E1 4NS, UK
3Department of Physics, Chungbuk National University, Cheongju, Chungbuk 28644, Korea
(Dated: March 24, 2026)
Traditional percolation theory assumes static microscopic rules, limiting its ability to describe
real-world complex systems where macroscopic order actively regulates local interactions. Here,
we introduce feedback percolation, an unified framework that dynamically couples the microscopic
activation probability to the macroscopic size of the giant component. In this work, we introduce feedback percolation models
in which the activation probability of links depends dynamically on the size of the giant component. Our framework allows networks to continuously adapt its local connections in response to its own macroscopic connectivity.Where the paper says it
Where the paper says it
Dynamics of Feedback Percolation
The feedback mechanism is defined by the iterative update rule pn = p + f(Sn−1) pg1. The size of the giant component Sn is determined by this updated probability pn as in standard bond percolation pg1. p. 1 This leads
to an iterative update of the bond occupation probability and the corresponding size of the giant component. Specifically, at each step n, the activation probability pn
is given by
pn = p + fn(Sn−1), (1)
where fn(x) is the feedback function at step n, while the
size of the giant component Sn is uniquely determined by
the bond occupation probability pn as in standard bond
percolation [see Fig. We show that this simple feedback mechanism produces a rich variety of behaviors both analytically and numerically. Depending
on the feedback functions, the system exhibits explosive discontinuous jumps, hybrid transitions,
limit-cycle oscillations, and routes to chaos, absent in classical percolation. In this case, Sn is determined uniquely by pn that on its turn is determined
uniquely by Sn−1. Thus, these iterative equations can
be cast into a one-dimensional map [35], Sn = h(Sn−1),
with h(x) being independent of n.Where the paper says it
Where the paper says it
Where the paper says it
Theoretical Derivations
The authors derive self-consistency equations for feedback percolation models on random networks with a degree distribution P(k) pg3. The implicit iterative map with a one-step delayed feedback is given by un = 1 − p + f(1 − G0(un−1))1 − G1(un) pg3. p. 2 Combining Eqs. (1)–(3), we can obtain the implicit
iterative map with a one-step delayed feedback:
un = 1 −
p + f(1 − G0(un−1))1 − G1(un)
. Defining ϕ(x) = 1 − [p + f(1 − G0(x))] [1 − G1(x)], the
condition leads to ϕ
′
(1) = 1 where ϕ
′
(x) = dϕ(x)/dx. Therefore, the percolation threshold pc for the percolation transition satisfies the following condition pc+f(0) =
1
G′
1
(1) where G′
1
(1) = (⟨k
2
⟩ − ⟨k⟩)/⟨k⟩.Where the paper says it
Where the paper says it
Observed Behaviors and Transitions
The study analyzes diverse behaviors based on the feedback function f(S) pg4. Under positive feedback, the system exhibits a The increase of pn is accelerated with an increase in the
size of the giant component; once the jump occurs, the
system reaches the maximum value of S∞. The two transitions at pc and pd are driven by distinct mechanisms. Under positive feedback, the mutual reinforcement between the size of the giant component and local activation
accelerates the growth of the giant component, leading to
diverse transitions, including continuous, discontinuous,
and double transitions. Delayed negative feedback effectively regulates the growth of the connectivity, causing
the stable fixed point into stable oscillatory states. These
oscillatory dynamics provide a robust theoretical foundation for understanding self-regulating systems in the real
world. We also found that the non-monotonic feedback
function can cause chaotic behavior in the size of the giant component.discontinuous
increase in the size S∞ pg5. The two transitions at pc and pd are driven by distinct mechanisms pg5. p. 3Where the paper says it
Where the paper says it
Where the paper says it
Relationship to Other Models
The paper suggests a connection between feedback percolation and other dynamical percolation models pg1. The model can be directly recasted into the triadic percolation with negative regulations pg7. p. 6 This mapping suggests that
feedback dynamics on a single-layer network may provide an effective model to understand the catastrophic
collapse driven by inter-layer cascades. To place this mechanism in a broader context, we observe that this model can also be directly recasted into
the triadic percolation [20, 25, 26] with negative regulations, where the activity of the regulatory nodes associated to each link suppresses the probability that the link
is active on its turn. 5(b), we confirm that the numerical simulations based on our feedback mechanism (symbols) are
in great agreement with the theoretical predictions of cascading failures (lines), in which the transition point shows
the relation zpd ≈ 2. This mapping suggests that
feedback dynamics on a single-layer network may provide an effective model to understand the catastrophic
collapse driven by inter-layer cascades. The transition at pd also shows a hybrid transition, meaning that
both a discontinuous jump and a scaling behavior appear. We demonstrate that the feedback provides an alternative interpretation for cascading failures in interdependent networks.Where the paper says it
Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Dynamic Phase Transition Analysis: The improved AI system can analyze complex network systems by incorporating feedback percolation models to predict a rich variety of behaviors including
explosive discontinuous jumps, hybrid transitions, limit-cycle oscillations, and routes to chaos.
-
Self-Regulating System Modeling: The system can model self-regulating systems by using delayed negative feedback to show how
the growth of a macroscopic structure triggers its own suppression,
providing a robust theoretical foundation for understanding such dynamics. -
Chaotic System Identification: The AI can identify chaotic regimes in networked systems where the
route to chaos is in the logistic map universality class
by computing Lyapunov exponents, specifically identifying regions where lambda greater than zero. -
Cascading Failure Prediction: The system can predict cascading failures in interdependent networks by mapping feedback percolation to cascading failures on two-layer ER networks, suggesting that
feedback dynamics on a single-layer network may provide an effective model to understand the catastrophic collapse driven by inter-layer cascades.
-
Transition Characterization: The improved AI system can distinguish between different types of transitions, such as the
hybrid nature
of discontinuous jumps and scaling behavior in the positive feedback model, which is characterized by a critical exponent beta equal to 1/2.
Abstract
Traditional percolation theory assumes static microscopic rules, limiting its ability to describe real-world complex systems where macroscopic order actively regulates local interactions. Here, we introduce feedback percolation, an unified framework that dynamically couples the microscopic activation probability to the macroscopic size of the giant component. We show that this simple feedback mechanism produces a rich variety of behaviors both analytically and numerically. Depending on the feedback functions, the system exhibits explosive discontinuous jumps, hybrid transitions, limit-cycle oscillations, and routes to chaos, absent in classical percolation. Our findings establish that macroscopic feedback provides a unifying physical mechanism for phenomena ranging from self-regulating oscillations to systemic infrastructure collapse.
Sources
- Dynamical entanglement percolation with spatially correlated disorder
- Dynamics of feedback Ising model
- Mixed phases in feedback Ising models
Related papers
- Emergent long-tail dynamics in driven magnets with dynamical frustration
- Statistical Mechanics of Dynamical System Identification
- Second law of thermodynamics in closed quantum many-body systems
- Chemical potentials from structure factors: II. Charged multi-component mixtures
- Genuine quantum scars in Floquet chaotic many-body systems
- Stochastic Thermodynamics for Autoregressive Generative Models: A Non-Markovian Perspective