arXiv subjects

Weiran Cai

Publications and source records attributed to Weiran Cai.

At least 19 recordsLinked to original sources

arXiv2026

Decoding species coexistence: A reinforcement learning perspective

A central goal in ecology is to understand how biodiversity is maintained. Previous theoretical works have employed the rock-paper-scissors (RPS) game as a toy model, demonstrating that population mobility is crucial in determining the species' coexistence. One key prediction is that biodiversity is jeopardized and eventually lost when mobility exceeds a certain value--a conclusion at odds with empirical observations of highly mobile species coexisting in nature. To address this discrepancy, we introduce a reinforcement learning framework and study a spatial RPS model, where individual mobility is adaptively regulated via a Q-learning algorithm rather than held fixed. Our results show that all three species can coexist stably, with extinction probabilities remaining low across a broad range of baseline migration rates. Mechanistic analysis reveals that individuals develop two behavioral tendencies: survival priority (escaping from predators) and predation priority (remaining near prey). While species coexistence emerges from the balance of the two tendencies, their imbalance jeopardizes biodiversity. Notably, there is a symmetry-breaking of action preference in a particular state that is responsible for the divergent species densities. Furthermore, when Q-learning species interact with fixed-mobility counterparts, those with adaptive mobility exhibit a significant evolutionary advantage. Our study suggests that reinforcement learning may offer a promising new perspective for uncovering the mechanisms of biodiversity and informing conservation strategies.

q-bio.PE↗

arXiv2025

Optimal coordination of resources: A solution from reinforcement learning

Efficient allocation is important in nature and human society, where individuals frequently compete for limited resources. The Minority Game (MG) is perhaps the simplest toy model to address this issue. However, most previous solutions assume that the strategies are provided a priori and static, failing to capture their adaptive nature. Here, we introduce the reinforcement learning (RL) paradigm to MG, where individuals adjust decisions based on accumulated experience and expected rewards dynamically. We find that this RL framework achieves optimal resource coordination when individuals balance the exploitation of experience with random exploration. Yet, the imbalanced strategies of the two lead to suboptimal partial coordination or even anti-coordination. Our mechanistic analysis reveals a symmetry-breaking in action preferences at the optimum, offering a fresh solution to the MG and new insights into the resource allocation problem.

physics.soc-ph↗

arXiv2025

Evolution of cooperation with Q-learning: the impact of information perception

The inherent complexity of human beings manifests in a remarkable diversity of responses to intricate environments, enabling us to approach problems from varied perspectives. However, in the study of cooperation, existing research within the reinforcement learning framework often assumes that individuals have access to identical information when making decisions, which contrasts with the reality that individuals frequently perceive information differently. In this study, we employ the Q-learning algorithm to explore the impact of information perception on the evolution of cooperation in a two-person Prisoner's Dilemma game. We demonstrate that the evolutionary processes differ significantly across three distinct information perception scenarios, highlighting the critical role of information structure in the emergence of cooperation. Notably, the asymmetric information scenario reveals a complex dynamical process, including the emergence, breakdown, and reconstruction of cooperation, mirroring psychological shifts observed in human behavior. Our findings underscore the importance of information structure in fostering cooperation, offering new insights into the establishment of stable cooperative relationships among humans.

q-bio.PE↗

arXiv2025

Evolution of cooperation in a bimodal mixture of conditional cooperators

Extensive behavioral experiments reveal that conditional cooperation is a prevalent phenomenon. Previous game-theoretical studies have predominantly relied on hard-manner models, where cooperation is triggered only upon reaching a specific threshold. However, this approach contrasts with the observed flexibility of human behaviors, where individuals adapt their strategies dynamically based on their surroundings. To capture this adaptability, we introduce a soft form of conditional cooperation by integrating the Q-learning algorithm from reinforcement learning. In this form, players not only reciprocate mutual cooperation but may also defect in highly cooperative environments or cooperate in less cooperative settings to maximize rewards. To explore the effects of hard and soft conditional cooperators, we examine their interactions in two scenarios: structural mixture (SM) and probabilistic mixture (PM), where the two behavioral modes are fixed and probabilistically adopted, respectively. In SM, hard conditional cooperators enhance cooperation when the threshold is low but hinder it otherwise. Surprisingly, in PM, the cooperation prevalence exhibits two first-order phase transitions as the probability is varied, leading to high, low, and vanishing levels of cooperation. Analysis of Q-tables offers insights into the "psychological shifts" of soft conditional cooperators and the overall evolutionary dynamics. Model extensions confirm the robustness of our findings. These results highlight the novel complexities arising from the diversity of conditional cooperators.

q-bio.PE↗

arXiv2024

Mitigating Exposure Bias in Score-Based Generation of Molecular Conformations

Molecular conformation generation poses a significant challenge in the field of computational chemistry. Recently, Diffusion Probabilistic Models (DPMs) and Score-Based Generative Models (SGMs) are effectively used due to their capacity for generating accurate conformations far beyond conventional physics-based approaches. However, the discrepancy between training and inference rises a critical problem known as the exposure bias. While this issue has been extensively investigated in DPMs, the existence of exposure bias in SGMs and its effective measurement remain unsolved, which hinders the use of compensation methods for SGMs, including ConfGF and Torsional Diffusion as the representatives. In this work, we first propose a method for measuring exposure bias in SGMs used for molecular conformation generation, which confirms the significant existence of exposure bias in these models and measures its value. We design a new compensation algorithm Input Perturbation (IP), which is adapted from a method originally designed for DPMs only. Experimental results show that by introducing IP, SGM-based molecular conformation models can significantly improve both the accuracy and diversity of the generated conformations. Especially by using the IP-enhanced Torsional Diffusion model, we achieve new state-of-the-art performance on the GEOM-Drugs dataset and are on par on GEOM-QM9. We provide the code publicly at https://github.com/jia-975/torsionalDiff-ip.

cs.LG↗

arXiv2024

Multiplex Graph Contrastive Learning with Soft Negatives

Graph Contrastive Learning (GCL) seeks to learn nodal or graph representations that contain maximal consistent information from graph-structured data. While node-level contrasting modes are dominating, some efforts commence to explore consistency across different scales. Yet, they tend to lose consistent information and be contaminated by disturbing features. Here, we introduce MUX-GCL, a novel cross-scale contrastive learning paradigm that utilizes multiplex representations as effective patches. While this learning mode minimizes contaminating noises, a commensurate contrasting strategy using positional affinities further avoids information loss by correcting false negative pairs across scales. Extensive downstream experiments demonstrate that MUX-GCL yields multiple state-of-the-art results on public datasets. Our theoretical analysis further guarantees the new objective function as a stricter lower bound of mutual information of raw input features and output embeddings, which rationalizes this paradigm. Code is available at https://github.com/MUX-GCL/Code.

cs.LG↗

arXiv2024

Hybrid Focal and Full-Range Attention Based Graph Transformers

The paradigm of Transformers using the self-attention mechanism has manifested its advantage in learning graph-structured data. Yet, Graph Transformers are capable of modeling full range dependencies but are often deficient in extracting information from locality. A common practice is to utilize Message Passing Neural Networks (MPNNs) as an auxiliary to capture local information, which however are still inadequate for comprehending substructures. In this paper, we present a purely attention-based architecture, namely Focal and Full-Range Graph Transformer (FFGT), which can mitigate the loss of local information in learning global correlations. The core component of FFGT is a new mechanism of compound attention, which combines the conventional full-range attention with K-hop focal attention on ego-nets to aggregate both global and local information. Beyond the scope of canonical Transformers, the FFGT has the merit of being more substructure-aware. Our approach enhances the performance of existing Graph Transformers on various open datasets, while achieves compatible SOTA performance on several Long-Range Graph Benchmark (LRGB) datasets even with a vanilla transformer. We further examine influential factors on the optimal focal length of attention via introducing a novel synthetic dataset based on SBM-PATTERN.

cs.LG↗

arXiv2024

Evolution of cooperation in the public goods game with Q-learning

Recent paradigm shifts from imitation learning to reinforcement learning (RL) is shown to be productive in understanding human behaviors. In the RL paradigm, individuals search for optimal strategies through interaction with the environment to make decisions. This implies that gathering, processing, and utilizing information from their surroundings are crucial. However, existing studies typically study pairwise games such as the prisoners' dilemma and employ a self-regarding setup, where individuals play against one opponent based solely on their own strategies, neglecting the environmental information. In this work, we investigate the evolution of cooperation with the multiplayer game -- the public goods game using the Q-learning algorithm by leveraging the environmental information. Specifically, the decision-making of players is based upon the cooperation information in their neighborhood. Our results show that cooperation is more likely to emerge compared to the case of imitation learning by using Fermi rule. Of particular interest is the observation of an anomalous non-monotonic dependence which is revealed when voluntary participation is further introduced. The analysis of the Q-table explains the mechanisms behind the cooperation evolution. Our findings indicate the fundamental role of environment information in the RL paradigm to understand the evolution of cooperation, and human behaviors in general.

q-bio.PE↗

arXiv2024

Catalytic evolution of cooperation in a population with behavioural bimodality

The remarkable adaptability of humans in response to complex environments is often demonstrated by the context-dependent adoption of different behavioral modes. However, the existing game-theoretic studies mostly focus on the single-mode assumption, and the impact of this behavioral multimodality on the evolution of cooperation remains largely unknown. Here, we study how cooperation evolves in a population with two behavioral modes. Specifically, we incorporate Q-learning and Tit-for-Tat (TFT) rules into our toy model, where prisoner's dilemma game is played and we investigate the impact of the mode mixture on the evolution of cooperation. While players in Q-learning mode aim to maximize their accumulated payoffs, players within TFT mode repeat what their neighbors have done to them. In a structured mixing implementation where the updating rule is fixed for each individual, we find that the mode mixture greatly promotes the overall cooperation prevalence. The promotion is even more significant in the probabilistic mixing, where players randomly select one of the two rules at each step. Finally, this promotion is robust when players are allowed to adaptively choose the two modes by real-time comparison. In all three scenarios, players within the Q-learning mode act as catalyzer that turns the TFT players to be more cooperative, and as a result drive the whole population to be highly cooperative. The analysis of Q-tables explains the underlying mechanism of cooperation promotion, which captures the ``psychologic evolution" in the players' mind. Our study indicates that the variety of behavioral modes is non-negligible, and could be crucial to clarify the emergence of cooperation in the real world.

q-bio.PE↗

arXiv2024

Boosting MLPs with a Coarsening Strategy for Long-Term Time Series Forecasting

Deep learning methods have been exerting their strengths in long-term time series forecasting. However, they often struggle to strike a balance between expressive power and computational efficiency. Resorting to multi-layer perceptrons (MLPs) provides a compromising solution, yet they suffer from two critical problems caused by the intrinsic point-wise mapping mode, in terms of deficient contextual dependencies and inadequate information bottleneck. Here, we propose the Coarsened Perceptron Network (CP-Net), featured by a coarsening strategy that alleviates the above problems associated with the prototype MLPs by forming information granules in place of solitary temporal points. The CP-Net utilizes primarily a two-stage framework for extracting semantic and contextual patterns, which preserves correlations over larger timespans and filters out volatile noises. This is further enhanced by a multi-scale setting, where patterns of diverse granularities are fused towards a comprehensive prediction. Based purely on convolutions of structural simplicity, CP-Net is able to maintain a linear computational complexity and low runtime, while demonstrates an improvement of 4.1% compared with the SOTA method on seven forecasting benchmarks.

cs.LG↗

arXiv2023

Decoding trust: A reinforcement learning perspective

Behavioral experiments on the trust game have shown that trust and trustworthiness are universal among human beings, contradicting the prediction by assuming \emph{Homo economicus} in orthodox Economics. This means some mechanism must be at work that favors their emergence. Most previous explanations however need to resort to some factors based upon imitative learning, a simple version of social learning. Here, we turn to the paradigm of reinforcement learning, where individuals update their strategies by evaluating the long-term return through accumulated experience. Specifically, we investigate the trust game with the Q-learning algorithm, where each participant is associated with two evolving Q-tables that guide one's decision making as trustor and trustee respectively. In the pairwise scenario, we reveal that high levels of trust and trustworthiness emerge when individuals appreciate both their historical experience and returns in the future. Mechanistically, the evolution of the Q-tables shows a crossover that resembles human's psychological changes. We also provide the phase diagram for the game parameters, where the boundary analysis is conducted. These findings are robust when the scenario is extended to a latticed population. Our results thus provide a natural explanation for the emergence of trust and trustworthiness without external factors involved. More importantly, the proposed paradigm shows the potential in deciphering many puzzles in human behaviors.

q-bio.PE↗

arXiv2023

A Generative Hypergraph Model for Double Heterogeneity

While network science has become an indispensable tool for studying complex systems, the conventional use of pairwise links often shows limitations in describing high-order interactions properly. Hypergraphs, where each edge can connect more than two nodes, have thus become a new paradigm in network science. Yet, we are still in lack of models linking network growth and hyperedge expansion, both of which are commonly observable in the real world. Here, we propose a generative hypergraph model by employing the preferential attachment mechanism in both nodes and hyperedge formation. The model can produce bi-heterogeneity, exhibiting scale-free distributions in both hyperdegree and hyperedge size. We provide a mean-field treatment that gives the expression of the two scaling exponents, which agree with the numerical simulations. Our model may help to understand the networked systems showing both types of heterogeneity and facilitate the study of complex dynamics thereon.

physics.soc-ph↗

arXiv2022

Degree-targeted cascades in modular, degree-heterogeneous networks

The dynamics of cascading activation, such as rapid changes in public opinion and the outbreak of disease epidemics, have a crucial dependence on the connectivity patterns among the agents. We study cascading dynamics in modular, degree-heterogeneous networks, and consider the impact of intra-module seeding strategy on inter-module spread. Specifically, we establish that although activating the highest-degree nodes is more effective than random selection at growing a cascade locally, there is a critical level of inter-module connectivity required for a cascade to cross from one module to another, irrespective of the seeding strategy. We present an analytical proof of this statement for the case that each module has the same degree distribution and all module pairs have the same inter-module connectivity, while our simulation results suggest its validity for more general situations, including a ring of modules. Interestingly, we find that on a network comprised of two modules, this critical level is primarily determined by the degree distribution of the \emph{alter} module, as opposed to the seed module. Our analytical approach extends a method developed by Gleeson, but is able to capture different seeding strategies using only one dynamical variable per module, namely the conditional exposure probability. Our work shows that the possibility of a global cascade depends sensitively on inter-module connectivity, and less on the intra-module seeding strategy. This suggests, for example, that slight changes to inter-module connectivity can be a feasible intervention strategy to promote or inhibit global cascades.

physics.soc-ph↗

arXiv2021

Detecting chaos in hurricane intensity

Determining the maximum potential limit in the accuracy of hurricane intensity prediction is important for operational practice. Using the phase-space reconstruction method for hurricane intensity time series, here we found that hurricane dynamics contain inherent low-dimensional chaos at the maximum intensity equilibrium. Examination of several chaotic invariants including the largest Lyapunov exponent, the Sugihara-May correlation, and the correlation dimension consistently captures an intrinsic dimension of the hurricane chaotic attractor in the range of 4-5. In addition, the error doubling time is roughly 1-5 hours, which accords with the decay time obtained from the Sugihara-May correlation. The confirmation of hurricane chaotic intensity as found in this study suggests a relatively short limit for intensity predictability of $\sim$18-24 hours after reaching the maximum intensity stage. So long as the traditional metrics for hurricane intensity such as the maximum surface wind or the minimum central pressure is used for intensity forecast, our results support that hurricane intensity forecast errors will not be reduced indefinitely in any modelling systems, even in the absence of all model and observational errors. As such, the future improvement of hurricane intensity forecast should be based on different intensity metric beyond the absolute intensity errors as in the current practice of intensity verification.

physics.ao-ph↗

arXiv2021

Quantifying the Global Support Network for Non-State Armed Groups (NAGs)

Human history has been shaped by armed conflicts. Rather than large-scale interstate wars, low-intensity attacks have been more prevalent in the post-World War era. These attacks are often carried out by non-state armed groups (NAGs), which are supported by host states (HSs). We analyze the global bipartite network of NAG-HS support and its evolution over the period of 1945-2010. We find striking parallels to ecological networks such as mutualistic and parasitic forms of support, and a nested and modular network architecture. The nestedness emerges from preferential behaviors: highly connected players are more likely to both gain and lose connections. Long-persisting major modules are identified, reflecting both regional and trans-regional interests, which show significant turnover in their membership, contrary to the transitory ones. Revealing this architecture further enables the identification of actor's roles and provide insights for effective intervention strategies.

cs.SI↗

arXiv2018

A Dynamic Niche Model for the Emergence and Evolution of Mutualistic Network Structures

Mutualistic interactions are vital constituents of ecological and socio-economic systems. Empirical studies have found that the patterns of reciprocal relations among the participants often shows the salient features of being simultaneously nested and modular. Whether and how these two structural properties of mutualistic networks can emerge out of a common mechanism however remains unclear. We propose a unified dynamic model based on the adaptation of niche relations that gives rise to both structural features. We apply Hutchinson's concept of niche interaction to networked cooperative species. Their niche relation evolves under the assumption of fitness maximization. Modularity and nestedness emerge concurrently through the accumulated local advantages in the structural and demographic distribution. A rich ensemble of key dynamical behaviors are unveiled in the dynamical framework. We demonstrate that mutualism can exhibit either a stabilizing or destabilizing effect on the evolved network, which undergoes a drastic transition with the overall competition level. Most strikingly, the adaptive network may exhibit a profound nature of history-dependency in response to environmental changes, allowing it to be found in alternative stable structures. The adaptive nature of niche interactions, as captured in our framework, can underlie a broad class of ecological relations and also socio-economic networks that engage in bipartite cooperation.

q-bio.PE↗

arXiv2015

Phase Transitions in Cooperative Coinfections: Simulation Results for Networks and Lattices

We study the spreading of two mutually cooperative diseases on different network topologies, and with two microscopic realizations, both of which are stochastic versions of an SIR type model studied by us recently in mean field approximation. There it had been found that cooperativity can lead to first-order spreading/extinction transitions. However, due to the rapid mixing implied by the mean field assumption, first order transitions required non-zero initial densities of sick individuals. For the stochastic model studied here the results depend strongly on the underlying network. First order transitions are found when there are few short but many long loops: (i) No first order transitions exist on trees and on 2-d lattices with local contacts (ii) They do exist on Erdos-Renyi (ER) networks, on d-dimensional lattices with d >= 4, and on 2-d lattices with sufficiently long-ranged contacts; (iii) On 3-d lattices with local contacts the results depend on the microscopic details of the implementation; (iv) While single infected seeds can always lead to infinite epidemics on regular lattices, on ER networks one sometimes needs finite initial densities of infected nodes; (v) In all cases the first order transitions are actually "hybrid", i.e. they display also power law scaling usually associated with second order transitions. On regular lattices, our model can also be interpreted as the growth of an interface due to cooperative attachment of two species of particles. Critically pinned interfaces in this model seem to be in different universality classes than standard critically pinned interfaces in models with forbidden overhangs. Finally, the detailed results mentioned above hold only when both diseases propagate along the same network of links. If they use different links, results can be rather different in detail, but are similar overall.

physics.soc-ph↗

arXiv2013

Outbreaks of coinfections: the critical role of cooperativity

Modeling epidemic dynamics plays an important role in studying how diseases spread, predicting their future course, and designing strategies to control them. In this letter, we introduce a model of SIR (susceptible-infected-removed) type which explicitly incorporates the effect of {\it cooperative coinfection}. More precisely, each individual can get infected by two different diseases, and an individual already infected with one disease has an increased probability to get infected by the other. Depending on the amount of this increase, we observe different threshold scenarios. Apart from the standard continuous phase transition for single disease outbreaks, we observe continuous transitions where both diseases must coexist, but also discontinuous transitions are observed, where a finite fraction of the population is already affected by both diseases at the threshold. All our results are obtained in a mean field model using rate equations, but we argue that they should hold also in more general frameworks.

q-bio.PE↗