Search arXiv⌕ Search

arXiv subjects

Jiaqi Zhu

Publications and source records attributed to Jiaqi Zhu.

60 records · Page 4Linked to original sources

Revisiting Heterophily For Graph Neural Networks

Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using graph structures based on the relational inductive bias (homophily assumption). While GNNs have been commonly believed to outperform NNs in real-world tasks, recent work has identified a non-trivial set of datasets where their performance compared to NNs is not satisfactory. Heterophily has been considered the main cause of this empirical observation and numerous works have been put forward to address it. In this paper, we first revisit the widely used homophily metrics and point out that their consideration of only graph-label consistency is a shortcoming. Then, we study heterophily from the perspective of post-aggregation node similarity and define new homophily metrics, which are potentially advantageous compared to existing ones. Based on this investigation, we prove that some harmful cases of heterophily can be effectively addressed by local diversification operation. Then, we propose the Adaptive Channel Mixing (ACM), a framework to adaptively exploit aggregation, diversification and identity channels node-wisely to extract richer localized information for diverse node heterophily situations. ACM is more powerful than the commonly used uni-channel framework for node classification tasks on heterophilic graphs and is easy to be implemented in baseline GNN layers. When evaluated on 10 benchmark node classification tasks, ACM-augmented baselines consistently achieve significant performance gain, exceeding state-of-the-art GNNs on most tasks without incurring significant computational burden.

cs.LG↗

Novel total hip surgery robotic system based on self-localization and optical measurement

This paper presents the development and experimental evaluation of a surgical robotic system for total hip arthroplasty (THA). Although existing robotic systems used in joint replacement surgery have achieved some progresses, the robot arm must be situated accurately at the target position during operation, which depends significantly on the experience of the surgeon. In addition, handheld acetabulum reamers typically exhibit uneven strength and grinding file. Moreover, the lack of techniques to real-time measure femoral neck length may lead to poor outcomes. To tackle these challenges, we propose a real-time traceable optical positioning strategy to reduce unnecessary manual adjustments to the robotic arm during surgery, an end-effector system to stabilise grinding, and an optical probe to provide real-time measurement of the femoral neck length and other parameters used to choose the proper prosthesis. The lengths of the lower limbs are measured as the prosthesis is installed. The experimental evaluation results show that, based on its accuracy, execution ability, and robustness, the proposed surgical robotic system is feasible for THA.

cs.RO↗

Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?

Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using the graph structures based on the relational inductive bias (homophily assumption). Though GNNs are believed to outperform NNs in real-world tasks, performance advantages of GNNs over graph-agnostic NNs seem not generally satisfactory. Heterophily has been considered as a main cause and numerous works have been put forward to address it. In this paper, we first show that not all cases of heterophily are harmful for GNNs with aggregation operation. Then, we propose new metrics based on a similarity matrix which considers the influence of both graph structure and input features on GNNs. The metrics demonstrate advantages over the commonly used homophily metrics by tests on synthetic graphs. From the metrics and the observations, we find some cases of harmful heterophily can be addressed by diversification operation. With this fact and knowledge of filterbanks, we propose the Adaptive Channel Mixing (ACM) framework to adaptively exploit aggregation, diversification and identity channels in each GNN layer to address harmful heterophily. We validate the ACM-augmented baselines with 10 real-world node classification tasks. They consistently achieve significant performance gain and exceed the state-of-the-art GNNs on most of the tasks without incurring significant computational burden.

cs.LG↗

Inference for Network Structure and Dynamics from Time Series Data via Graph Neural Network

Network structures in various backgrounds play important roles in social, technological, and biological systems. However, the observable network structures in real cases are often incomplete or unavailable due to measurement errors or private protection issues. Therefore, inferring the complete network structure is useful for understanding complex systems. The existing studies have not fully solved the problem of inferring network structure with partial or no information about connections or nodes. In this paper, we tackle the problem by utilizing time series data generated by network dynamics. We regard the network inference problem based on dynamical time series data as a problem of minimizing errors for predicting future states and proposed a novel data-driven deep learning model called Gumbel Graph Network (GGN) to solve the two kinds of network inference problems: Network Reconstruction and Network Completion. For the network reconstruction problem, the GGN framework includes two modules: the dynamics learner and the network generator. For the network completion problem, GGN adds a new module called the States Learner to infer missing parts of the network. We carried out experiments on discrete and continuous time series data. The experiments show that our method can reconstruct up to 100% network structure on the network reconstruction task. While the model can also infer the unknown parts of the structure with up to 90% accuracy when some nodes are missing. And the accuracy decays with the increase of the fractions of missing nodes. Our framework may have wide application areas where the network structure is hard to obtained and the time series data is rich.

cs.LG↗

Improved Exponential Time Lower Bound of Knapsack Problem under BT model

M.Alekhnovich et al. recently have proposed a model of algorithms, called BT model, which covers Greedy, Backtrack and Simple Dynamic Programming methods and can be further divided into fixed, adaptive and fully adaptive three kinds, and have proved exponential time lower bounds of exact and approximation algorithms under adaptive BT model for Knapsack problem which are $Ω(2^{n/2}/\sqrt n)=Ω(2^{0.5n}/\sqrt n)$ and $Ω((1/ε)^{1/3.17})\approxΩ((1/ε)^{0.315})$(for approximation ratio $1-ε$) respectively (M. Alekhovich, A. Borodin, J. Buresh-Oppenheim, R. Impagliazzo, A. Magen, and T. Pitassi, Toward a Model for Backtracking and Dynamic Programming, \emph{Proceedings of Twentieth Annual IEEE Conference on Computational Complexity}, pp308-322, 2005). In this note, we slightly improved their lower bounds to $Ω(2^{(2-ε)n/3}/\sqrt{n})\approx Ω(2^{0.66n}/\sqrt{n})$ and $Ω((1/ε)^{1/2.38})\approxΩ((1/ε)^{0.420})$, and proposed as an open question what is the best achievable lower bounds for knapsack under adaptive BT models.

cs.CC↗

On Lower Bound of Worst Case Error Probability for Quantum Fingerprinting with Shared Entanglement

This paper discusses properties of quantum fingerprinting with shared entanglement. Under certain restriction of final measurement, a relation is given between unitary operations of two parties. Then, by reducing to spherical coding problem, this paper gives a lower bound of worst case error probability for quantum fingerprinting with shared entanglement, showing a relation between worst case error probability and the amount of entanglement(measured by Schmidt number).

quant-ph↗