Author name cluster

Ao Zhou

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

1 author row

EAAI Journal 2026 Journal Article

Hybrid machine learning and physical modeling framework for climate-driven risk zonation of concrete shrinkage damage

Qiaosong Hu
Dujian Zou
Zhilin Bai
Tiejun Liu
Ao Zhou

Concrete shrinkage under non-stationary climatic forcing poses an increasing threat to the serviceability and longevity of infrastructure in low-pressure, arid and high-altitude regions. Current models neglect multi-environment interactions and climate-driven risk evolution. This study presents a hybrid modeling and assessment framework that coupled physics-informed empirical priors with optimized machine learning to predict shrinkage evolution, quantify structural risk, and map spatiotemporal vulnerability under future climate scenarios. A curated shrinkage database was fused with high-resolution meteorological projections and downscaled via filtering and cubic interpolation. The empirical CEB-FIP 2010 shrinkage formulation and air pressure parameters were embedded into feature engineering to create temperature-humidity-pressure coupled predictors. An XGBoost (Extreme Gradient Boosting) model was optimized through systematic hyperparameter tuning and physics-guided transfer learning. The optimized coupling model attained R2 = 0. 92 to predict shrinkage evolution, and reduced long-term prediction divergence to within 15% against independent data from three-factor experiments. To translate material-level shrinkage into structural risk, multiphysics finite-element simulations of a representative reinforced-concrete pier incorporated eigenstrain shrinkage fields and reinforcement constraint to resolve strain–stress–damage progression. Four critical normalized strain thresholds were identified that demarcated initiation, stable propagation, accelerated expansion and through-crack stages. A five-tier risk zoning map across China was constructed, covering both historical data and mid-future climate scenario. Plateau and northwestern basins showed marked vulnerability. Using C60 concrete as a representative case study due to its prevalence, results showed the medium-to-high risk area increasing by 65%, with 31. 1% of China's territory classified as medium–high risk by 2050.

Details DOI

AAAI Conference 2026 Conference Paper

SlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning

Lingkun Long
Rubing Yang
Yushi Huang
Desheng Hui
Ao Zhou
Jianlei Yang

Long-context inference for Large Language Models (LLMs) is heavily limited by high computational demands. While several existing methods optimize attention computation, they still process the full set of hidden states at each layer, limiting overall efficiency. In this work, we propose SlimInfer, an innovative framework that aims to accelerate inference by directly pruning less critical prompt tokens during the forward pass. Our key insight is an information diffusion phenomenon: As information from critical tokens propagates through layers, it becomes distributed across the entire sequence. This diffusion process suggests that LLMs can maintain their semantic integrity when excessive tokens, even including these critical ones, are pruned in hidden states. Motivated by this, SlimInfer introduces a dynamic fine-grained pruning mechanism that accurately removes redundant tokens of hidden state at intermediate layers. This layer-wise pruning naturally enables an asynchronous KV cache manager that prefetches required token blocks without complex predictors, reducing both memory usage and I/O costs. Extensive experiments show that SlimInfer can achieve up to 2.53× time-to-first-token (TTFT) speedup and 1.88× end-to-end latency reduction for LLaMA3.1-8B-Instruct on a single RTX 4090, without sacrificing performance on LongBench.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Batch Selection for Multi-Label Classification Guided by Uncertainty and Dynamic Label Correlations

Ao Zhou
Bin Liu
Jin Wang
Grigorios Tsoumakas

The accuracy of deep neural networks is significantly influenced by the effectiveness of mini-batch construction during training. In single-label scenarios, such as binary and multi-class classification tasks, it has been demonstrated that batch selection algorithms preferring samples with higher uncertainty achieve better performance than difficulty-based methods. Although there are two batch selection methods tailored for multi-label data, none of them leverage important uncertainty information. Adapting the concept of uncertainty to multi-label data is not a trivial task, since there are two issues that should be tackled. First, traditional variance or entropy-based uncertainty measures ignore fluctuations of predictions within sliding windows and the importance of the current model state. Second, existing multi-label methods do not explicitly exploit the label correlations, particularly the uncertainty-based label correlations that evolve during the training process. In this paper, we propose an uncertainty-based multi-label batch selection algorithm. It assesses uncertainty for each label by considering differences between successive predictions and the confidence of current outputs, and further leverages dynamic uncertainty-based label correlations to emphasize instances whose uncertainty is synergistically expressed across multiple labels. Empirical studies demonstrate the effectiveness of our method in improving the performance and accelerating the convergence of various multi-label deep learning models.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

CHPO: Constrained Hybrid-action Policy Optimization for Reinforcement Learning

Ao Zhou
Jiayi Guan
Li Shen
Fan Lu
Sanqing Qu
Junqiao Zhao
Ziqiao Wang
Ya Wu

Constrained hybrid-action reinforcement learning (RL) promises to learn a safe policy within a parameterized action space, which is particularly valuable for safety-critical applications involving discrete-continuous hybrid action spaces. However, existing hybrid-action RL algorithms primarily focus on reward maximization, which faces significant challenges for tasks involving both cost constraints and hybrid action spaces. In this work, we propose a novel Constrained Hybrid-action Policy Optimization algorithm (CHPO) to address the problems of constrained hybrid-action RL. Concretely, we rethink the limitations of hybrid-action RL in handling safe tasks with parameterized action spaces and reframe the objective of constrained hybrid-action RL by introducing the concept of Constrained Parameterized-action Markov Decision Process (CPMDP). Subsequently, we present a constrained hybrid-action policy optimization algorithm to confront the constrained hybrid-action problems and conduct theoretical analyses demonstrating that the CHPO converges to the optimal solution while satisfying safety constraints. Finally, extensive experiments demonstrate that the CHPO achieves competitive performance across multiple experimental tasks.

PDF Details

EAAI Journal 2025 Journal Article

Django-based framework database for leakage detection using machine learning for water distribution networks

Yiwei Xie
Mengze Gao
Fan Luo
Ao Zhou
Yunfeng Yang
Jian Hu
Wei Jiang
Yuanyao Ye

Leakage in water supply pipe networks is a critical issue, with traditional detection methods being inefficient and error-prone. Acoustic-based leak detection often lacks standardized databases, limiting its effectiveness. This study proposes an integrated system using MySQL, Python, and Django for managing and analyzing acoustic leakage data. The system incorporates Variable Modal Decomposition (VMD), Wavelet Threshold Noise Reduction, Feature Extraction, and Support Vector Machine (SVM) for accurate leak detection. Experimentation on 413 labeled acoustic samples achieved classification accuracies of 96. 1% (training set) and 97. 4% (test set). This approach enhances detection precision and offers a scalable solution for real-time monitoring, with significant practical implications for improving water distribution system management and decision-making.

Details DOI

NeurIPS Conference 2023 Conference Paper

VOCE: Variational Optimization with Conservative Estimation for Offline Safe Reinforcement Learning

Jiayi Guan
Guang Chen
Jiaming Ji
Long Yang
Ao Zhou
Zhijun Li
Changjun Jiang

Offline safe reinforcement learning (RL) algorithms promise to learn policies that satisfy safety constraints directly in offline datasets without interacting with the environment. This arrangement is particularly important in scenarios with high sampling costs and potential dangers, such as autonomous driving and robotics. However, the influence of safety constraints and out-of-distribution (OOD) actions have made it challenging for previous methods to achieve high reward returns while ensuring safety. In this work, we propose a Variational Optimization with Conservative Eestimation algorithm (VOCE) to solve the problem of optimizing safety policies in the offline dataset. Concretely, we reframe the problem of offline safe RL using probabilistic inference, which introduces variational distributions to make the optimization of policies more flexible. Subsequently, we utilize pessimistic estimation methods to estimate the Q-value of cost and reward, which mitigates the extrapolation errors induced by OOD actions. Finally, extensive experiments demonstrate that the VOCE algorithm achieves competitive performance across multiple experimental tasks, particularly outperforming state-of-the-art algorithms in terms of safety.

PDF Details