Author name cluster

Tao Wen

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.

3 papers

1 author row

TCS Journal 2026 Journal Article

Extra path-structure connectivity of modified bubble-sort networks

Guozhen Zhang
Tao Wen
Dajin Wang

A network’s connectivity is a crucial indicator for its reliability. There are various ways to measure the connectivity, and the extra connectivity and the structure connectivity are two variants of the classic, original connectivity. In this paper, we incorporate the two to study the extra structure connectivity for the modified bubble-sort network MBn, which is one of the proposed models for the interconnection network of multiprocessor systems. Let H be a connected subgraph of a graph G, and let F = { H 1, H 2, …, H j } be a set of subgraphs of G, such that 1) each Hi is isomorphic to H; 2) G − F is disconnected; and 3) each component of G − F has at least g + 1 nodes. The minimum j for such an F is called the g-extra H-structure connectivity of G, denoted κg (G; H). Let F = { J 1, J 2, …, J k } be a set of subgraphs of G, such that 1) each Ji is isomorphic to a subgraph of H; 2) G − F is disconnected; and 3) each component of G − F has at least g + 1 nodes. The minimum k for such an F is called the g-extra H-substructure connectivity of G, denoted κ g s ( G; H ). We will prove that for P 3l, a path on 3l nodes, κ 1 ( M B n; P 3 l ) = κ 1 s ( M B n; P 3 l ) = ⌈ n − 1 l ⌉ for n ≥ 9 and l ≤ n − 2.

Details DOI

EAAI Journal 2024 Journal Article

Data-driven hierarchical learning approach for multi-point servo control of Pan–Tilt–Zoom cameras

Haitao Wang
XiangShuai Zhai
Tao Wen
ZiDu Yin
Yang Yang

Pan–Tilt–Zoom (PTZ) cameras, with their significant features of free rotation and zoom, are widely used in areas such as border security, ecological conservation, emergency management, and the military. PTZ cameras can achieve automatic monitoring of a selected area through multiple servo operations, known as multi-point servo control. However, due to the deficiencies in the servo control Software Development Kit (SDK), hardware wear, and interference from complex external environments, the multi-point servo control process generates significant errors. This paper proposes a precise multi-point servo control framework based on Deep Reinforcement Learning (DRL) to address this issue. The complexity of real-world environments necessitates a reward function design that fully considers various factors, for which we propose the directional gravity reward function. Due to the instability during the training process, prolonged trial-and-error interactions between the agent and the equipment can cause irreversible damage to the devices. This framework employs a phased training approach, where the agent learns sequentially from offline, off-policy, and on-policy data, reducing direct interaction with the equipment while enhancing the agent’s overall performance. Additionally, real-time and accurate device status can be obtained by performing feature matching on images from adjacent time frames, which is crucial for the system’s operation. Evaluation results indicate that our proposed precise multi-point servo control framework significantly outperforms other methods in 4-point servo control tasks in both virtual and real-world scenarios. Additionally, the operational process fully considers safety issues.

Details DOI

IJCAI Conference 2021 Conference Paper

Knowledge-based Residual Learning

Guanjie Zheng
Chang Liu
Hua Wei
Porter Jenkins
Chacha Chen
Tao Wen
Zhenhui Li

Small data has been a barrier for many machine learning tasks, especially when applied in scientific domains. Fortunately, we can utilize domain knowledge to make up the lack of data. Hence, in this paper, we propose a hybrid model KRL that treats domain knowledge model as a weak learner and uses another neural net model to boost it. We prove that KRL is guaranteed to improve over pure domain knowledge model and pure neural net model under certain loss functions. Extensive experiments have shown the superior performance of KRL over baselines. In addition, several case studies have explained how the domain knowledge can assist the prediction.

PDF Details DOI