arXiv subjects

Bo Jiao

Publications and source records attributed to Bo Jiao.

6 recordsLinked to original sources

arXiv2026

Ollivier-Ricci curvature in cycle overlap mode

Ollivier-Ricci curvature of an edge (x,y) is defined by comparing the distance taken to transport from neighbors of x to neighbors of y. It is a structural measure that has been studied in many fields such as community detection and deep neural networks. However, high computational complexity or error limits its application on large scale-free graphs. This paper proposes an optimal transport principle to minimize the distance by 3,4,5-cycles that include the edge (x,y), and designs a curvature calculation approach named Curvature in Cycle Overlap Mode (CCOM). In this approach, a greedy and pruning algorithm is proposed to approximate the optimal transport principle. We theoretically and experimentally verified that our approach CCOM can significantly improve the accuracy of the curvature on real-world networks with low time consumption. In addition, we compared CCOM with baseline approximation approaches in community detection tasks using the same curvature-based framework, and experimentally confirmed the effectiveness of CCOM on large scale-free graphs. Open source code is available at https://github.com/hitreasonx/CCOM.

cs.SI↗

arXiv2025

Hierarchical graph sampling based minibatch learning with chain preservation and variance reduction

Graph sampling-based Graph Convolutional Networks (GCNs) decouple sampling from forward and backward propagation during minibatch training, enhancing scalability with respect to layer depth and graph size. We propose HIS_GCNs, a hierarchical importance sampling-based learning method. By constructing minibatches using sampled subgraphs, HIS_GCNs focuses on the importance of both the core and periphery in a scale-free training graph. Specifically, it preserves the centrum of the core in most minibatches, which maintains connectivity between periphery nodes, and samples periphery edges without core node interference, which allows longer chains composed entirely of low-degree nodes remain within the same minibatch. HIS_GCNs can maximize the discrete Ricci curvature (i.e., Ollivier-Ricci curvatures) of the edges in a subgraph, enabling preservation of important chains for information propagation. This approach can achieve a low node embedding variance and a high convergence speed. Diverse experiments on Graph Neural Networks (GNNs) with node classification tasks confirmed the superior performance of HIS_GCNs in terms of both accuracy and training time. Open-source code (https://github.com/HuQiaCHN/HIS-GCN).

cs.LG↗

arXiv2023

Sampling unknown large networks restricted by low sampling rates

Graph sampling plays an important role in data mining for large networks. Specifically, larger networks often correspond to lower sampling rates. Under the situation, traditional traversal-based samplings for large networks usually have an excessive preference for densely-connected network core nodes. Aim at this issue, this paper proposes a sampling method for unknown networks at low sampling rates, called SLSR, which first adopts a random node sampling to evaluate a degree threshold, utilized to distinguish the core from periphery, and the average degree in unknown networks, and then runs a double-layer sampling strategy on the core and periphery. SLSR is simple that results in a high time efficiency, but experimental evaluation confirms that the proposed method can accurately preserve many critical structures of unknown large networks with low sampling rates and low variances.

cs.DS↗

arXiv2022

CA-SpaceNet: Counterfactual Analysis for 6D Pose Estimation in Space

Reliable and stable 6D pose estimation of uncooperative space objects plays an essential role in on-orbit servicing and debris removal missions. Considering that the pose estimator is sensitive to background interference, this paper proposes a counterfactual analysis framework named CASpaceNet to complete robust 6D pose estimation of the spaceborne targets under complicated background. Specifically, conventional methods are adopted to extract the features of the whole image in the factual case. In the counterfactual case, a non-existent image without the target but only the background is imagined. Side effect caused by background interference is reduced by counterfactual analysis, which leads to unbiased prediction in final results. In addition, we also carry out lowbit-width quantization for CA-SpaceNet and deploy part of the framework to a Processing-In-Memory (PIM) accelerator on FPGA. Qualitative and quantitative results demonstrate the effectiveness and efficiency of our proposed method. To our best knowledge, this paper applies causal inference and network quantization to the 6D pose estimation of space-borne targets for the first time. The code is available at https://github.com/Shunli-Wang/CA-SpaceNet.

cs.CV↗

arXiv2021

Harvesting the triplet excitons of quasi-two-dimensional perovskite toward highly efficient white light-emitting diodes

Utilization of triplet excitons, which generally emit poorly, is always fundamental to realize highly efficient organic light-emitting diodes (LEDs). While triplet harvest and energy transfer via electron exchange between triplet donor and acceptor are fully understood in doped organic phosphorescence and delayed fluorescence systems, the utilization and energy transfer of triplet excitons in quasi-two-dimensional (quasi-2D) perovskite are still ambiguous. Here, we use an orange-phosphorescence-emitting ultrathin organic layer to probe triplet behavior in the sky-blue-emitting quasi-2D perovskite. The delicate white LEDs architecture enables a carefully tailored Dexter-like energy-transfer mode that largely rescues the triplet excitons in quasi-2D perovskite. Our white organic-inorganic LEDs achieve maximum forward-viewing external quantum efficiency of 8.6% and luminance over 15000 cd m-2, exhibiting a significant efficiency enhancement versus the corresponding sky-blue perovskite LED (4.6%). The efficient management of energy transfer between excitons in quasi-2D perovskite and Frenkel excitons in organic layer opens the door to fully utilizing excitons for white organic-inorganic LEDs.

physics.optics↗

arXiv2020

Impermeable Inorganic Walls Sandwiching Photoactive Layer toward Inverted Perovskite Solar and Indoor-Photovoltaic Devices

Interfaces between the perovskite active layer and the charge-transport layers (CTLs) play a critical role in both efficiency and stability of halide-perovskite photovoltaics. One of the major concerns is that surface defects of perovskite could cause detrimental nonradiative recombination and material degradation. In this work, we addressed this challenging problem by inserting ultrathin alkali-fluoride (AF) films between the tri-cation lead-iodide perovskite layer and both CTLs. This bilateral inorganic walls strategy makes use of both physical-blocking and chemical-anchoring functionalities of the continuous, uniform and compact AF framework: on the one hand, the uniformly distributed alkali-iodine coordination at the perovskite-AF interfaces effectively suppresses the formation of iodine-vacancy defects at the surfaces and grain boundaries of the whole perovskite film, thus reducing the trap-assisted recombination at the perovskite-CTL interfaces and therewith the open-voltage loss; on the other hand, the impermeable AF buffer layers effectively prevent the bidirectional ion migration at the perovskite-CTLs interfaces even under harsh working conditions. As a result, a power-conversion efficiency (PCE) of 22.02% (certified efficiency 20.4%) with low open-voltage deficit (< 0.4V) was achieved for the low-temperature processed inverted planar perovskite solar cells. Exceptional operational stability (500 h, ISOS-L-2) and thermal stability (1000 h, ISOS-D-2) were obtained. Meanwhile, a 35.7% PCE was obtained under dim-light source (1000 lux white LED light) with the optimized device, which is among the best records in perovskite indoor photovoltaics.

cond-mat.mtrl-sci↗