Dynamic Policy Research Articles

Boosting Weak-to-Strong Agents in Multiagent Reinforcement Learning via Balanced PPO.

Multiagent policy gradients (MAPGs), an essential branch of reinforcement learning (RL), have made great progress in both industry and academia. However, existing models do not pay attention to the inadequate training of individual policies, thus limiting the overall performance. We verify the existence of imbalanced training in multiagent tasks and formally define it as an imbalance between policies (IBPs). To address the IBP issue, we propose a dynamic policy balance (DPB) model to balance the learning of each policy by dynamically reweighting the training samples. In addition, current methods for better performance strengthen the exploration of all policies, which leads to disregarding the training differences in the team and reducing learning efficiency. To overcome this drawback, we derive a technique named weighted entropy regularization (WER), a team-level exploration with additional incentives for individuals who exceed the team. DPB and WER are evaluated in homogeneous and heterogeneous tasks, effectively alleviating the imbalanced training problem and improving exploration efficiency. Furthermore, the experimental results show that our models can outperform the state-of-the-art MAPG methods and boast over 12.1 % performance gain on average.

Blockchain-Aided Anonymous Traceable and Revocable Access Control Scheme With Dynamic Policy Updating for the Cloud IoT

The Internet of Things has been applied in various fields of industry, which has promoted the intelligent development of the industry and improved efficiency in industrial production. The devices involved in the IoT have generated useful and sensitive data over time and upload the data to the cloud to realize real-time data sharing. To ensure the confidentiality of data, many systems use attribute-based encryption primitive to encrypt data. However, there are still some security and privacy problems in this mode, such as the lack of identification of malicious users who leaked private keys, performance bottleneck caused by excessive reliance on a single central authority, and vulnerability because a single central authority holds the private keys of all users in the system. In this paper, white-box tracking is used to identify malicious users. The alliance chain is introduced to support multi-authority environments, where the consensus nodes are managed by different authorities and assist central authority in generating partial private keys. To protect users’ privacy, users remain anonymous at all times during their interactions with blockchain consensus. The security analysis and simulation results show that the proposed scheme outperformed other comparable schemes, indicating that it is a preferable scheme.

Simple Monotonic Readjustment Policies with Applications to Markdown Pricing and Pricing in the Presence of Strategic Customers

A Simple Monotonic Dynamic Pricing Policy We consider a canonical revenue management setting in which a monopolist seeks to maximize expected total revenues by selling a fixed inventory of product to customers who arrive sequentially over time. It is known in the literature that a simple dynamic pricing policy can perform reasonably well in such setting. However, the literature typically assumes that price can move freely over time (i.e., it can either go up or down). In practice, firms often impose certain restrictions on price movement. In this paper, we consider the setting in which price can only move monotonically over time (either only decreasing or only increasing). We devise a simple provably near-optimal dynamic pricing policy. Aside from markdown applications, our policy can also be applied to the setting with strategic customers.

Climate change in open environments: Revisiting the current distribution to understand and safeguard the future of psammophilous squamates of the Diagonal of Open Formations of South America

To identify regions that contribute to the persistence of biodiversity, we (1) predict the potential distribution of the psammophilous squamate species endemic of the Diagonal of Open Formations (DOF) in the current climate, (2) identify survey priority areas, (3) estimate the impacts of climate change in two future CO2 emission scenarios, and (4) discuss strategies to safeguard these target species. We created an occurrence database and used environmental variables for the current and future periods to run the potential distribution models. We obtained models with high performance for all species and scenarios. Our study indicates potential occurrence areas for the psammophilous squamates of the DOF that still need to be sampled and are a survey priority for future inventories. In all future projections, climate change is predicted to shift the suitability and promote species range change, modifying the richness patterns for the psammophilous squamate reptiles of DOF. Finally, we demonstrate that future climatic conditions may decrease the effectiveness of current areas in protecting the diversity of these reptiles. Our results highlight the vulnerability of psammophilous squamates from South American Dry Diagonal to climate change and provide crucial information of the need of dynamic public policies for conservation of these species.

Trade-off or synergy? Dynamic analysis and policy insights on land use functions in China

Limited land resources and diverse utilization have resulted in complex trade-offs and synergies (TS) among land use functions (LUFs). Analyzing these TS and their spatio-temporal variations is crucial for developing effective land use policies to regulate and manage multiple LUFs, promoting efficient land resource utilization and fostering sustainable development in society, economy, and ecology. This study uses interdisciplinary research methods, including the projection pursuit model, Spearman's rank correlation analysis, and exploratory spatial data analysis, to investigate the characteristics of LUFs TS in 31 provincial administrative divisions in China from 2009 to 2020. Key findings include: (1) LUFs exhibited significant spatial heterogeneity and temporal variations. Most LUFs experienced distinct upgrading processes, except for SSF (Social Security Function), which requires specific attention.(2) LUFs TS were found to be common and consistent, with varying characteristics observed in different years. It is crucial to consider these characteristics during the process of functional regulation to enhance effectiveness. (3) We confirmed that LUFs, as a public service, show substantial spatial spillover effects, resulting in cross-regional LUFs TS. Regulatory strategies should harness the radiating and propelling effects of advantageous functions while maintaining equilibrium and coordination in regional development, ultimately achieving optimal overall benefits. Finally, we discuss the study findings in comparison with existing research and provide policy insights to promote the harmonized and sustainable development of LUFs. These findings aim to serve as a scientific foundation for land use planning and management in China and offer valuable insights for other regions.

Dynamic modeling and policy simulation to reduce heat-related illness risk from urban heatwaves in Seoul, South Korea

Integrated and systematic studies on heat-related illness risk from urban heatwaves are lacking. The effects and predictions of risk reduction policies for heat-related illness risk should not be limited to a particular area of study but should be calculated in a feedback process considering the budget. In this study, we performed dynamic modeling and policy simulation aimed at reducing the risk of heat-related illnesses caused by urban heatwaves in Seoul, Korea. A causal map of the urban heatwave system was created using data from 2010 to 2021, and simulations were performed up to 2040. In analyzing the effect of reducing the incidence of patients with heat-related illnesses compared to the budget invested, the number of patients with heat-related illnesses to be reduced in support policies for the vulnerable group, high-efficiency air conditioners, heatwave shelters, green areas, and eco-friendly cars would be reduced by 0.44, 0.09, 0.0046, 0.0045, and 0.0005 people/billion won in 2035, respectively. The number of heat-related illnesses decreased by 78.0% when policy simulation, which produced the maximum effect with the lowest budget, was performed. The current urban heatwave system in Seoul would not reduce the incidence of patients with heat-related illnesses in the future but, rather, would increase it. Urban heatwave policies did not contribute considerably to the reduction of patients with heat-related illnesses because the intensity of heatwaves owing to climate change outweighed the impact of heatwave policies. We suggest additional budget increases in the order of higher return on investment (ROI).

Prevalence and network structure of post-traumatic stress symptoms and their association with suicidality among Chinese mental health professionals immediately following the end of China’s Dynamic Zero-COVID Policy: a national survey

Studies on post-traumatic stress symptoms (PTSS) among mental health professionals (MHPs) are limited, particularly since restrictions due to coronavirus disease (COVID-19) have been lifted such as the recent termination of China’s Dynamic Zero-COVID Policy. The current study filled this gap by exploring the prevalence, correlates, and network structure of PTSS as well as its association with suicidality from a network analysis perspective. A cross-sectional, national survey was conducted using a convenience sampling method on MHPs between January 22 and February 10, 2023. PTSS were assessed using the Post-Traumatic Stress Disorder Checklist-Civilian version, while suicidality was assessed using standardized questions related to ideation, plans, and attempts. Univariate and multivariate analyses examined correlates of PTSS. Network analysis explored the structure of PTSS and suicidality. The centrality index of “Expected influence” was used to identify the most central symptoms in the network, reflecting the relative importance of each node in the network. The “flow” function was adopted to identify specific symptoms that were directly associated with suicidality. A total of 10,647 MHPs were included. The overall rates of PTSS and suicidality were 6.7% (n = 715; 95% CI = 6.2–7.2%) and 7.7% (n = 821; 95% CI = 7.2–8.2%), respectively. Being married (OR = 1.523; P < 0.001), quarantine experience (OR = 1.288; P < 0.001), suicidality (OR = 3.750; P < 0.001) and more severe depressive symptoms (OR = 1.229; P < 0.001) were correlates of more PTSS. Additionally, higher economic status (e.g., good vs. poor: OR = 0.324; P = 0.001) and health status (e.g., good vs. poor: OR = 0.456; P < 0.001) were correlates of reduced PTSS. PCL6 (“Avoiding thoughts”; EI = 1.189), PCL7 (“Avoiding reminders”; EI = 1.157), and PCL11 (“Feeling emotionally numb”; EI = 1.074) had the highest centrality, while PCL12 (“Negative belief”), PCL 16 (“Hypervigilance”) and PCL 14 (“Irritability”) had the strongest direct, positive associations with suicidality. A high prevalence of lingering PTSS was found among MHPs immediately after China’s “Dynamic Zero-COVID Policy” was terminated. Avoidance and hyper-arousal symptoms should be monitored among at-risk MHPs after the COVID-19 pandemic and serve as potential targets for the prevention and treatment of PTSS in this population.

Kebijakan Moneter dan Transformasinya

Monetary policy is a vital instrument in managing a country's economic stability. Global economic transformation and dynamics in financial markets require in-depth evaluation of monetary policy. This research aims to analyze the influencing factors of monetary policy and investigate the impact of its transformation in the current economic context. Through a careful methodological approach, this research uses secondary data analysis from various global and national economic sources. The literature review investigates the current literature on monetary policy and economic transformation. The results of the data analysis provide deep insight into the factors influencing monetary policy and the consequences of its transformation. The main findings show that changes in the global economic order, financial technology and capital market dynamics play a major role in changing the monetary policy paradigm. This transformation can have a significant impact on economic stability and societal welfare. The conclusions of this research provide a basis for developing policies that are responsive and adaptive to economic transformation. The practical implications of this research include the need to update a more dynamic monetary policy strategy to answer new challenges in the era of globalization. It is hoped that further research will deepen specific aspects of monetary policy transformation to provide a more comprehensive view.

Relationship between housing, oil, gold and stock markets: evidence from UK and Norway

Purpose The purpose of this study is to explore the intricate relationship between oil prices, house prices in the UK and Norway, and the mediating role of gold and stock prices in both the short- and long-term, unraveling these complex linkages by employing an empirical approach. Design/methodology/approach This study benefits from a comprehensive set of econometric tools, including a multiequation vector autoregressive (VAR) system, Granger causality test, impulse response function, variance decomposition and a single-equation autoregressive distributed lag (ARDL) system. This rigorous approach enables to identify both short- and long-run dynamics to unravel the intricate linkages between Brent oil prices, housing prices, gold prices and stock prices in the UK and Norway over the period from 2005:Q1 to 2022:Q2. Findings The findings indicate that rising oil prices negatively impact house prices, whereas the positive influence of stock market performance on housing is more pronounced. A two-way causal relationship exists between stock market indices and house prices, whereas a one-way causal relationship exists from crude oil prices to house prices in both countries. The VAR model reveals that past housing prices, stock market indices in each country and Brent oil prices are the primary determinants of current housing prices. The single-equation ARDL results for housing prices demonstrate the existence of a long-run cointegrating relationship between real estate and stock prices. The variance decomposition analysis indicates that oil prices have a more pronounced impact on housing prices compared with stock prices. The findings reveal that shocks in stock markets have a greater influence on housing market prices than those in oil or gold prices. Consequently, house prices exhibit a stronger reaction to general financial market indicators than to commodity prices. Research limitations/implications This study may have several limitations. First, the model does not include all relevant macroeconomic variables, such as interest rates, unemployment rates and gross domestic product growth. This omission may affect the accuracy of the model’s predictions and lead to inefficiencies in the real estate market. Second, this study does not consider alternative explanations for market inefficiencies, such as behavioral finance factors, information asymmetry or market microstructure effects. Third, the models have limitations in revealing how predictors react to positive and negative shocks. Therefore, the results of this study should be interpreted with caution. Practical implications These findings hold significant implications for formulating dynamic policies aimed at stabilizing the housing markets of these two oil-producing nations. The practical implications of this study extend to academics, investors and policymakers, particularly in light of the volatility characterizing both housing and commodity markets. The findings reveal that shocks in stock markets have a more profound impact on housing market prices compared with those in oil or gold prices. Consequently, house prices exhibit a stronger reaction to general financial market indicators than to commodity prices. Social implications These findings could also serve as valuable insights for future research endeavors aimed at constructing models that link real estate market dynamics to macroeconomic indicators. Originality/value Using a variety of econometric approaches, this paper presents an innovative empirical analysis of the intricate relationship between euro property prices, stock prices, gold prices and oil prices in the UK and Norway from 2005:Q1 to 2022:Q2. Expanding upon the existing literature on housing market price determinants, this study delves into the role of gold and oil prices, considering their impact on industrial production and overall economic growth. This paper provides valuable policy insights for effectively managing the impact of oil price shocks on the housing market.

Using discrete-event simulation to assess an AHP-based dynamic patient prioritisation policy for elective surgery

ABSTRACTPatient prioritisation in elective surgery is a critical multi-criteria decision-making issue. Elective patients should be ranked on the waiting list based on various criteria related to their health conditions and their evolution over time. This study suggests a dynamic prioritisation where elective patients are ranked considering their evolving health conditions, waiting time, and patient loss, i.e., the likelihood of patients abandoning the waiting list. We proposed deploying the Analytic Hierarchy Process (AHP) to weigh prioritisation criteria. The AHP-derived weights are integrated into a discrete-event simulation model, which periodically ranks patients considering their health status, waiting time, and patient loss. The proposed dynamic prioritisation is compared with the first-come, first-served (FCFS) policy and the static prioritisation, which ranks patients according to their initial priority. Results showed that dynamic prioritisation outperforms the FCFS and static prioritisation policies in preventing health deterioration, reducing weighted waiting time, and reducing the ratio of lost patients.

The optimisation of public health emergency governance: a simulation study based on COVID-19 pandemic control policy

BackgroundThe outbreak of the COVID-19 pandemic sparked numerous studies on policy options for managing public health emergencies, especially regarding how to choose the intensity of prevention and control to maintain a balance between economic development and disease prevention.MethodsWe constructed a cost-benefit model of COVID-19 pandemic prevention and control policies based on an epidemic transmission model. On this basis, numerical simulations were performed for different economies to analyse the dynamic evolution of prevention and control policies. These economies include areas with high control costs, as seen in high-income economies, and areas with relatively low control costs, exhibited in upper-middle-income economies.ResultsThe simulation results indicate that, at the outset of the COVID-19 pandemic, both high-and low-cost economies tended to enforce intensive interventions. However, as the virus evolved, particularly in circumstances with relatively rates of reproduction, short incubation periods, short spans of infection and low mortality rates, high-cost economies became inclined to ease restrictions, while low-cost economies took the opposite approach. However, the consideration of additional costs incurred by the non-infected population means that a low-cost economy is likely to lift restrictions as well.ConclusionsThis study concludes that variations in prevention and control policies among nations with varying income levels stem from variances in virus transmission characteristics, economic development, and control costs. This study can help researchers and policymakers better understand the differences in policy choice among various economies as well as the changing trends of dynamic policy choices, thus providing a certain reference value for the policy direction of global public health emergencies.

Fully dynamic reorder policies with deep reinforcement learning for multi-echelon inventory management

The operation of inventory systems plays an important role in the success of manufacturing companies, making it a highly relevant domain for optimization. In particular, the domain lends itself to being approached via Deep Reinforcement Learning (DRL) models due to it requiring sequential reorder decisions based on uncertainty to minimize cost. In this paper, we evaluate state-of-the-art optimization approaches to determine whether Deep Reinforcement Learning can be applied to the multi-echelon inventory optimization (MEIO) framework in a practically feasible manner to generate fully dynamic reorder policies. We investigate how it performs in comparison to an optimized static reorder policy, how robust it is when it comes to structural changes in the environment, and whether the use of DRL is safe in terms of risk in real-world applications. Our results show promising performance for DRL with potential for improvement in terms of minimizing risky behavior.

Depression and Anxiety Symptoms and Their Associated Factors Among Chinese Residents After the Lifting of the Dynamic Zero-COVID Policy: A Cross-Sectional Study.

China lifted its strict zero-Covid approach on December 7, 2022. This study aimed to investigate depression and anxiety symptoms and their associations among Chinese residents after the change in public policy. A cross-sectional sample of 925 Chinese residents (726 females and 199 males) was recruited using convenience and snowball sampling approach between 16 and 25 December 2022. Participants completed online questionnaires on basic information, depression, anxiety, COVID-19 related perceptions, and protective behaviors change. Mild and moderate-to-severe depression symptoms were reported by 35.6% and 19.1% of participants, respectively. Nearly 40% of participants reported mild anxiety and 18.7% reported moderate-to-severe anxiety. Results from multinomial logistic regression analysis indicated that male gender, younger age, the presence of chronic disease, poorer self-rated mental health status, perceived impact, and worry were risk factors for both depression and anxiety, while higher education and protective behaviors change were protective factors. Besides, living with or caring for children (4-6 years), family members or other housemates currently with influenza-like symptoms, and perceived severity were also risk factors for depression. Our findings provided initial evidence that Chinese residents may face heightened depression and anxiety during the early stage after the policy was released. Furthermore, we identified some vulnerable populations in need of prioritizing mental health assistance and some potentially modifiable factors associated with depression and anxiety, which provides an important guide for developing timely and effective psychological interventions and preparing for future pandemics.

Inferring Complementary and Substitutable Products Based on Knowledge Graph Reasoning

Complementarity and substitutability between products are essential concepts in retail and marketing. To achieve this, existing approaches take advantage of knowledge graphs to learn more evidence for inference. However, they often omit the knowledge that lies in the unstructured data. In this research, we concentrate on inferring complementary and substitutable products in e-commerce from mass structured and unstructured data. An improved knowledge-graph-based reasoning model has been proposed which cannot only derive related products but also provide interpretable paths to explain the relationship. The methodology employed in our study unfolds through several stages. First, a knowledge graph refining entities and relationships from data was constructed. Second, we developed a two-stage knowledge representation learning method to better represent the structured and unstructured knowledge based on TransE and SBERT. Then, the relationship inferring problem was converted into a path reasoning problem under the Markov decision process environment by learning a dynamic policy network. We also applied a soft pruning strategy and a modified reward function to improve the effectiveness of the policy network training. We demonstrate the effectiveness of the proposed method on standard Amazon datasets, and it gives about 5–15% relative improvement over the state-of-the-art models in terms of NDCG@10, Recall@10, Precision @10, and HR@10.

Dynamic Routing Policies for Multi-Skill Call Centers Using Deep Q Network

When the call center queuing system becomes complex, it turns out that the static routing policy is not optimal. This paper considers the problem of the dynamic routing policy for call centers with multiple skill types and agent groups. A state-dependent routing policy based on the Deep Q Network (DQN) is proposed, and a reinforcement learning algorithm is applied to optimize the routing. A simulation algorithm is designed to help customers and agents interact with the external environment to learn the optimal strategy. The performance evaluation considered in this paper is the service level/abandon rate. Experiments show that the DQN-based dynamic routing policy performs better than the common static policy Global First Come First Serve (FCFS) and the dynamic policy Priorities with Idle Agent Thresholds and Weight-Based Routing in various examples. On the other hand, the training time of the routing policy model based on the DQN is much faster than routing optimization based on simulation and a genetic algorithm.

Undernourishment on Longevity in West Africa: A Group Outcome

The length of time a person lives from the date of birth is an essential dimension of a country’s level of development. Several factors have been identified as responsible for longevity. This study investigated the effect of undernourishment prevalence and other factors on longevity in West Africa from 2000-2018. Using panel data, the random coefficient model was employed to examine the effect of a novel variable, such as the prevalence of undernourishment and others (government expenditure on health, population density, unemployment and out-of-pocket expenditure) on longevity. Except for a few, the findings revealed that longevity is a decreasing function of the prevalence of undernourishment in West Africa. In contrast, it was revealed that longevity is an increasing function of population density in all West African countries. The effect of government spending and unemployment on health contradicts theory in some countries, while they were in tandem with theory in other countries. Therefore, the study recommended, among others, the introduction of dynamic agricultural policies to increase income and ensure food security in West Africa.

Sustainable Water Policymaking for the Hotel Industry: A Longitudinal Network Analysis of Policy Documents

Safeguarding water resources in tourism contributes to the achievement of the United Nations Sustainable Development Goals (SDGs) and to the wider 2030 Agenda for Sustainable Development. However, sustainable management of water, in order to address both current and future demand for water resources in an equitable and environmentally sound manner, is one of the most significant challenges faced by tourist destinations. Using a case study approach, this research conducts a social network analysis (SNA) of the water policymaking system of Singapore’s hotel industry by longitudinally examining the structure of relations among policy themes and policy stakeholders. The network approach attempts to examine the structural evolution in the water demand management (WDM) policy network based on an analysis of 329 (co)produced policy documents (2001–2020) available in the public domain. A broad range of policy themes, together with divergent policy stakeholders, are uncovered in this interconnected dynamic policy system. Subsequently, systematic policy research is extended to discuss the interrelationships of policy stakeholders and policy themes over the past 20 years. Our findings highlight the network structure of the policy domain and suggest an openness and vibrancy of sustainable WDM as an interdisciplinary policy domain. Cross-sectoral stakeholder engagement has contributed to the evolution and complexity of the policy theme network. The study’s findings provide significant insights into the relationships of policy actors and the evolving network structure of a multidimensional policy framework.

Designing climate change dynamic adaptive policy pathways for agricultural water management using a socio-hydrological modeling approach

Deep climate uncertainties create challenges to long-term planning in the management of complicated water resources and agricultural systems. The use of an innovative decision-making approach under deep climate uncertainties is helpful to deal with these challenges in such systems. The research aims to develop an appropriate methodology for designing and evaluating robust adaptable plans in the agricultural sector in the face of deep climate uncertainties which covers hydrological aspects and stakeholders’ preferences and behaviors. For this purpose, Dynamic Adaptive Policy Pathways (DAPP) approach in conjunction with socio-hydrological modeling is utilized to design and evaluate comprehensive robust adaptable plans under future climate uncertainties in the Hablehroud River Basin, Iran. In this context, an agent-based modeling (ABM) approach is linked to the Soil and Water Assessment Tool (SWAT) model to account for the social behavior of various stakeholders as well as the impact of climate change on water resources and agricultural systems and their interactions. The results indicated that encouraging farmers to cooperate in changing cropping patterns and deficit irrigation should be proposed as a main plan in both short- and long-term planning. If the implementation issues are also considered, penalty-based adaptation actions (robustness range of 56–72 %) can be prioritized in the early years to gradually educate farmers about the benefits of cooperative behavior. Therefore, taking into account all issues: 1) sustainability of water resources, 2) agricultural income, 3) climate change impacts, and 4) implementation plans in the time period, it is suggested to impose a penalty on deficit irrigation (56–76 % of robustness) during the early time periods and then shift to changes in cropping patterns through training of farmers (76–82 %). Considering limits to allowable water withdrawal through penalties (56–72 %) can be implemented as another suggested alternative in the early time periods, but imposing penalties on deficit irrigation (62–68 %) should be implemented in the long-term.

Walking, the body, and the pandemic: the public value of walking art in China

ABSTRACT In December 2022, the dynamic zero-COVID control policy came to an end, marking the conclusion of a three-year pandemic that affected 1.4 billion Chinese people. The pandemic and related policies created a unique, temporary, and historic social ecosystem where walking became more crucial than ever before. The pandemic not only severely restricted people’s movement in public spaces but also exposed the longstanding contradictions between human bodies, modern mobility, and urban space. Over the three years of the pandemic, walking became an aesthetic survival attempt by Chinese people to cope with their limited freedoms under the pandemic. As the pandemic stagnated and worsened over time, walking-dominant activities gradually became a widespread social phenomenon that encouraged urban residents to participate in rebuilding society across various fields such as politics, art, nature, etc. The development of walking as an artistic form during this period represents a new aesthetic strategy and political awakening while reflecting humans’ need to reconnect with land, social space, and their own bodies. This paper reviews how walking art has evolved historically through three periods – before, during, and after the pandemic – aiming to highlight both the public value of walking art and challenges within China’s social ecosystem.

GWalloc: A self-adaptive generational wear-aware allocator for non-volatile main memory

Phase Change Memory (PCM) is considered a promising replacement for DRAM due to its superior performance characteristics such as low leakage power, high integration density, byte addressability and non-volatility. However, PCM’s limited write endurance significantly hinders its wide application. For example, PCM wears out quickly with traditional dynamic memory allocation policy in embedded systems which aggregates lots of writes in few memory blocks. To extend the lifespan of PCM, some wear-aware dynamic memory allocators have been proposed, which generally depend on some fixed parameters to limit the wear of PCM. However, these allocators can be inflexible as it is difficult to specify appropriate values for the required parameters in different scenarios. In this paper, we propose a Self-Adaptive Generational Wear-Aware Allocator (GWalloc). GWalloc divides memory blocks into two generations: the young and the old generation, according to their number of allocation times. GWalloc also dynamically adjusts the system’s wear threshold during allocations so that it can effectively balance the wear degree of PCM and the consumed memory space. The wear threshold restricts the upper wear limit of young memory blocks. Experimental evaluations show that compared with the state-of-the-art wear-aware dynamic memory allocators (NVMalloc, Walloc and UWLalloc), GWalloc improves PCM wear-leveling (evaluated by CV, a wear leveling indicator) by 38.6%, 39.1% and 38.3%, and saves 62.1%, 22.2% and 37.2% memory space overhead.