arXiv subjects

Wenzhen Huang

Publications and source records attributed to Wenzhen Huang.

5 recordsLinked to original sources

arXiv2023

How AI-driven Digital Twins Can Empower Mobile Networks

The growing complexity of next-generation networks exacerbates the modeling and algorithmic flaws of conventional network optimization methodology. In this paper, we propose a mobile network digital twin (MNDT) architecture for 6G networks. To address the modeling and algorithmic shortcomings, the MNDT uses a simulation-optimization structure. The feedback from the network simulation engine, which serves as validation for the optimizer's decision outcomes, is used explicitly to train artificial intelligence (AI) empowered optimizers iteratively. In practice, we develop a network digital twin prototype system leveraging data-driven technology to accurately model the behaviors of mobile network elements (e.g., mobile users and base stations), wireless environments, and network performance. An AI-powered network optimizer has been developed based on the deployed MNDT prototype system for providing reliable and optimized network configurations. The results of the experiments demonstrate that the proposed MNDT infrastructure can provide practical network optimization solutions while adapting to the more complex environment.

cs.NI↗

arXiv2023

Carbon emissions and sustainability of launching 5G mobile networks in China

Since 2021, China has deployed more than 2.1 million 5G base stations to increase the network capacity and provide ubiquitous digital connectivity for mobile terminals. However, the launch of 5G networks also exacerbates the misalignment between cellular traffic and energy consumption, which reduces carbon efficiency - the amount of network traffic that can be delivered for each unit of carbon emission. In this study, we develop a large-scale data-driven framework to estimate the carbon emissions induced by mobile networks. We show that the decline in carbon efficiency leads to a carbon efficiency trap, estimated to cause additional carbon emissions of 23.82 +- 1.07 megatons in China. To mitigate the misalignment and improve energy efficiency, we propose DeepEnergy, an energy-saving method leveraging collaborative deep reinforcement learning and graph neural networks. DeepEnergy models complex collaboration among cells, making it possible to effectively coordinate the working state of tens of thousands of cells, which could help over 71% of Chinese provinces avoid carbon efficiency traps. In addition, applying DeepEnergy is estimated to reduce 20.90 +- 0.98 megatons of carbon emissions at the national level in 2023. We further assess the effects of adopting renewable energy and discover that the mobile network could accomplish more than 50% of its net-zero goal by integrating DeepEnergy and solar energy systems. Our study provides insight into carbon emission mitigation in 5G network infrastructure launching in China and overworld, paving the way towards achieving sustainable development goals and future net-zero mobile networks.

eess.SY↗

arXiv2021

Planning with Exploration: Addressing Dynamics Bottleneck in Model-based Reinforcement Learning

Model-based reinforcement learning (MBRL) is believed to have higher sample efficiency compared with model-free reinforcement learning (MFRL). However, MBRL is plagued by dynamics bottleneck dilemma. Dynamics bottleneck dilemma is the phenomenon that the performance of the algorithm falls into the local optimum instead of increasing when the interaction step with the environment increases, which means more data can not bring better performance. In this paper, we find that the trajectory reward estimation error is the main reason that causes dynamics bottleneck dilemma through theoretical analysis. We give an upper bound of the trajectory reward estimation error and point out that increasing the agent's exploration ability is the key to reduce trajectory reward estimation error, thereby alleviating dynamics bottleneck dilemma. Motivated by this, a model-based control method combined with exploration named MOdel-based Progressive Entropy-based Exploration (MOPE2) is proposed. We conduct experiments on several complex continuous control benchmark tasks. The results verify that MOPE2 can effectively alleviate dynamics bottleneck dilemma and have higher sample efficiency than previous MBRL and MFRL algorithms.

cs.LG↗

arXiv2021

Learning to Reweight Imaginary Transitions for Model-Based Reinforcement Learning

Model-based reinforcement learning (RL) is more sample efficient than model-free RL by using imaginary trajectories generated by the learned dynamics model. When the model is inaccurate or biased, imaginary trajectories may be deleterious for training the action-value and policy functions. To alleviate such problem, this paper proposes to adaptively reweight the imaginary transitions, so as to reduce the negative effects of poorly generated trajectories. More specifically, we evaluate the effect of an imaginary transition by calculating the change of the loss computed on the real samples when we use the transition to train the action-value and policy functions. Based on this evaluation criterion, we construct the idea of reweighting each imaginary transition by a well-designed meta-gradient algorithm. Extensive experimental results demonstrate that our method outperforms state-of-the-art model-based and model-free RL algorithms on multiple tasks. Visualization of our changing weights further validates the necessity of utilizing reweight scheme.

cs.LG↗

arXiv2017

Transfer Learning based Dynamic Multiobjective Optimization Algorithms

One of the major distinguishing features of the dynamic multiobjective optimization problems (DMOPs) is the optimization objectives will change over time, thus tracking the varying Pareto-optimal front becomes a challenge. One of the promising solutions is reusing the "experiences" to construct a prediction model via statistical machine learning approaches. However most of the existing methods ignore the non-independent and identically distributed nature of data used to construct the prediction model. In this paper, we propose an algorithmic framework, called Tr-DMOEA, which integrates transfer learning and population-based evolutionary algorithm for solving the DMOPs. This approach takes the transfer learning method as a tool to help reuse the past experience for speeding up the evolutionary process, and at the same time, any population based multiobjective algorithms can benefit from this integration without any extensive modifications. To verify this, we incorporate the proposed approach into the development of three well-known algorithms, nondominated sorting genetic algorithm II (NSGA-II), multiobjective particle swarm optimization (MOPSO), and the regularity model-based multiobjective estimation of distribution algorithm (RM-MEDA), and then employ twelve benchmark functions to test these algorithms as well as compare with some chosen state-of-the-art designs. The experimental results confirm the effectiveness of the proposed method through exploiting machine learning technology.

cs.NE↗