The dynamic blocking caused by unmanned aerial vehicle (UAV) severely reduces the channel reliability of visible light communication (VLC), where conventional multipath diversity is unavailable. To address this, we propose an optical intelligent reflecting surface (OIRS)-assisted robust VLC framework against UAV blockage. To simulate and quantify blockage, we establish a unified channel-optical field-blocking model integrating: (i) a Gaussian-approximated optical power distribution for OIRS mirror units, derived from far-field-filtered Lambertian emissions; (ii) a probabilistic blockage model using Poisson point processes to quantify UAV-induced blockages. We formulate the optimization of light field distribution on OIRS as a dynamic optimization problem to maximize the user’s received power under blockage constraints by adjusting the deflection angle of VLC access point (AP). We propose a two-stage machine learning approach to solve this dynamic optimization problem. In the offline stage, we employ greedy combinatorial optimization to precompute diverse OIRS beam configurations for building a configuration database. In the online stage, we first detect occluded mirror units using sparse recovery, and then dynamically adjust the beam aiming center based on real-time conditions through database guidance selection. The simulation results show that our solution can significantly improve the actual anti-blocking performance under high-density mirror unit configuration compared to static state, random enable, and gradient descent.

A key challenge in decentralized optimization is determining the optimal convergence rate and designing algorithms to achieve it. While this problem has been extensively addressed for doubly-stochastic and column-stochastic mixing matrices, the row-stochastic scenario remains unexplored. This paper bridges this gap by introducing effective metrics to capture the influence of row-stochastic mixing matrices and establishing the first convergence lower bound for decentralized learning over row-stochastic networks. However, existing algorithms fail to attain this lower bound due to two key issues: deviation in the descent direction caused by the adapted gradient tracking (GT) and instability introduced by the Pull-Diag protocol. To address descent deviation, we propose a novel analysis framework demonstrating that Pull-Diag-GT achieves linear speedup—the first such result for row-stochastic decentralized optimization. Moreover, by incorporating a multi-step gossip (MG) protocol, we resolve the instability issue and attain the lower bound, achieving near-optimal complexity for decentralized optimization over row-stochastic networks.

The prosperity of massive open online courses provides fodder for plentiful research efforts on adaptive learning. However, current open-access educational datasets are still far from sufficient to meet the need for various topics of adaptive learning. Existing released datasets often cover only small-scale data, lack fine-grained knowledge concepts. They are even difficult to curate and supplement due to platform limitations. In this work, we construct MOOCCubeX, a large, knowledge-centered repository consisting of 4,216 courses, 230,263 videos, 358,265 exercises, 637,572 fine-grained concepts and over 296 million behavioral data of 3,330,294 students, for supporting the research topics on adaptive learning in MOOCs. Licensed by XuetangX, one of the largest MOOC websites in China, we obtain abundant and diverse course resources and student behavioral data and are permitted to make subsequent periodic updates. We propose a framework to accomplish data processing, weakly supervised fine-grained concept graph mining, and data curation to improve usability and richness. Based on the fine-grained concepts, we re-organize the data from the knowledge perspective and acquire more external learning resources from the web. Our repository is now available at https://github.com/THU-KEG/MOOCCubeX.

Jifan Yu, Gan Luo, Tong Xiao, Qingyang Zhong, Yuquan Wang, Wenzheng Feng, Junyi Luo, Chenyu Wang, Lei Hou, Juanzi Li, Zhiyuan Liu, Jie Tang. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020.

As Massive Open Online Courses (MOOCs) become increasingly popular, it is promising to automatically provide extracurricular knowledge for MOOC users.Suffering from semantic drifts and lack of knowledge guidance, existing methods can not effectively expand course concepts in complex MOOC environments.In this paper, we first build a novel boundary during searching for new concepts via external knowledge base and then utilize heterogeneous features to verify the highquality results.In addition, to involve human efforts in our model, we design an interactive optimization mechanism based on a game.Our experiments on the four datasets from Coursera 1 and XuetangX 2 show that the proposed method achieves significant improvements(+0.19by MAP) over existing methods.The source code 3 and datasets 4 have been published.

This paper presents WordRecorder, an efficient and accurate handwriting recognition system that identifies words using acoustic signals generated by pens and paper, thus enabling ubiquitous handwriting recognition. To achieve this, we carefully craft a new deep-learning based acoustic sensing framework with three major components, i.e., segmentation, classification, and word suggestion. First, we design a dual-window approach to segment the raw acoustic signal into a series of words and letters by exploiting subtle acoustic signal features of handwriting. Then we integrate a set of simple yet effective signal processing techniques to further refine raw acoustic signals into normalized spectrograms which are suitable for deep-learning classification. After that, we customize a deep neural network that is suitable for smart devices. Finally, we incorporate a word suggestion module to enhance the recognition performance. Our framework achieves both computation efficiency and desirable classification accuracy simultaneously. We prototype our design using off-the-shelf smartwatches and conduct extensive evaluations. Our results demonstrate that WordRecorder robustly archives 81% accuracy rate for trained users, and 75% for users without training, across a range of different environment, users, and writing habits.

Acoustic sensing has brought forth the advances of prosperous applications such as gesture recognition. Specifically, ambient acoustic sensing has drawn many contentions due to the ease of use property. Unfortunately, the inherent ambient noise is the major reason for unstable gesture recognition. In this work, we propose “SoundWrite II”, which is an improved version of our previously designed system. Compared with our previous design, we utilize the two threshold values to identify the effective signals from the original noisy input, and leverage the MFCC (Mel frequency cepstral coefficient) to extract the stable features from different gestures. These enhancements could effectively improve the noise tolerant performance for previous design. Implementation on the Android system has realized the real time processing of the feature extraction and gesture recognition. Extensive evaluations have validated our design, where the noise tolerant property is fully tested under different experimental settings and the recognition accuracy could be 91% with 7 typical gestures.