检索结果-内蒙古大学图书馆

Learning interpretable deep state space model for probabilistic time series forecasting

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Longyuan Yan, Junchi Yang, Xiaokang Jin, Yaohui State Key Lab of Advanced Optical Communication System and Network MoE Key Lab of Artificial Intelligence AI Institute Department of Computer Science and Engineering Shanghai Jiao Tong University China

Probabilistic time series forecasting involves estimating the distribution of future based on its history, which is essential for risk management in downstream decision-making. We propose a deep state space model for probabilistic time series forecasting whereby the non-linear emission model and transition model are parameterized by networks and the dependency is modeled by recurrent neural nets. We take the automatic relevance determination (ARD) view and devise a network to exploit the exogenous variables in addition to time series. In particular, our ARD network can incorporate the uncertainty of the exogenous variables and eventually helps identify useful exogenous variables and suppress those irrelevant for forecasting. The distribution of multi-step ahead forecasts are approximated by Monte Carlo simulation. We show in experiments that our model produces accurate and sharp probabilistic forecasts. The estimated uncertainty of our forecasting also realistically increases over time, in a spontaneous manner. © 2021, CC BY.

关键词： state space methods

High speed data encryption and decryption using stimulated Brillouin scattering effect in optical fiber

学校读者我要写书评

暂无评论

Proceedings of SPIE - The International Society for optical Engineering 2011年 8198卷

作者： Yi, Lilin Zhang, Tao Hu, Weisheng State Key Lab of Advanced Optical Communication System and Networks Shanghai Jiao Tong University Department of Electronic and Engineering Shanghai 200240 China

ISBN: (纸本)9780819488398

A novel all-optical encryption/decryption method based on stimulated Brillouin scattering (SBS) effect in optical fiber is proposed for the first time. The operation principle is explained in detail and the encryption and decryption performance is experimentally evaluated. The encryption keys could be the SBS gain amplitude, bandwidth, central wavelength and spectral shape, which are configurable and flexibly controlled by the users. We experimentally demonstrate the SBS encryption/decryption process of a 10.86-Gb/s non-return-to-zero (NRZ) data by using both phase-modulated and current-dithered Brillouin pumps for proof-of-concept. Unlike the traditional optical encryption methods of chaotic communications and optical code-division-multiplexing access (OCDMA), the SBS based encryption/decryption technique can directly upgrade the current optical communication system to a secure communication system without changing the terminal transceivers, which is completely compatible with the current optical communication systems. © 2011 SPIE.

关键词： Cryptography

End-to-end conversational search for online shopping with utterance transfer

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Xiao, Liqiang Ma, Jun Dong, Xin Luna Martinez-Gomez, Pascual Zalmout, Nasser Zhang, Chenwei Zhao, Tong He, Hao Jin, Yaohui MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University *** Inc State Key Lab of Advanced Optical Communication System and Network Shanghai Jiao Tong University

Successful conversational search systems can present natural, adaptive and interactive shopping experience for online shopping customers. However, building such systems from scratch faces real word challenges from both imperfect product schema/knowledge and lack of training dialog data. In this work we first propose ConvSearch, an end-to-end conversational search system that deeply combines the dialog system with search. It leverages the text profile to retrieve products, which is more robust against imperfect product schema/knowledge compared with using product attributes alone. We then address the lack of data challenges by proposing an utterance transfer approach that generates dialogue utterances by using existing dialog from other domains, and leveraging the search behavior data from e-commerce retailer. With utterance transfer, we introduce a new conversational search dataset for online shopping. Experiments show that our utterance transfer method can significantly improve the availability of training dialogue data without crowd-sourcing, and the conversational search system significantly outperformed the best tested baseline. © 2021, CC BY-NC-SA.

关键词： Electronic commerce

616 MHz repetition rate sub-100fs amplifier-similariton Yb:Fiber laser

学校读者我要写书评

暂无评论

616 MHz repetition rate sub-100fs amplifier-similariton Yb:F...

OptoElectronics and communications Conference, OECC

作者： Chen Li Guizhong Wang Tongxiao Jiang Peng Li Aimin Wang Zhigang Zhang State Key Lab State Key Laboratory of Advanced Optical Communication System and Networks Peking University Beijing P.R. China

We report a compact Yb:fiber ring laser operated at 616 MHz. The average output power is 450 mW at a pump power of 900 mW. The dechirped pulse width is 98 fs.

关键词： optical fiber amplifiers Fiber lasers Cavity resonators Laser mode locking optical fiber dispersion optical fiber polarization

Distortion-aware 2D soft decision for VCSEL-MMF optical PAM interconnection

学校读者我要写书评

暂无评论

Distortion-aware 2D soft decision for VCSEL-MMF optical PAM ...

optical Fiber communication Conference, OFC 2020

作者： Sun, Lin Du, Jiangbing Zhang, Wenjia Chi, Nan Lu, Chao He, Zuyuan State Key Laboratory of Advanced Optical Communication Systems and Networks Shanghai Jiao Tong University Shanghai200240 China Department of Electronic and Information Engineering Photonics Research Centre Hong Kong Polytechnic University Hung Hom Kowloon Hong Kong Department of Communication Science and Engineering State Key Lab of ASIC and System Fudan University Shanghai200433 China

A distortion-aware 2D soft decision method of PAM signals have been proposed for VCSEL-MMF interconnection system. Improvements and application potential have been experimentally investigated on a 112-Gbps optical PAM... 详细信息

ISBN: (数字)9781943580712

ISBN: (纸本)9781943580712

关键词： optical fiber communication

Micro-Bending Influence on Mode-Dependent Loss

学校读者我要写书评

暂无评论

Micro-Bending Influence on Mode-Dependent Loss

International Conference on advanced Infocomm Technology (ICAIT)

作者： Junpeng Liang Gang Qiao Jinlong Wei Juhao Li Yu Yang Zhaopeng Xu Lulu Liu Shangcheng Wang Qi Wu Weisheng Hu Dept. of Circuits and System Peng Cheng Laboratory Shenzhen China State Key Lab of Advanced Optical Communication Systems and Network Peking University Beijing China

Micro-bending with random curvature is usually used to model the mode coupling in few mode fiber (FMF). However, previous models overlooked the accompanied mode dependent loss (MDL) and neglected the order constraints of coupling loss between different modes. In this work, we calculate out the accompanied loss of each mode by using the finite element method. The results show that MDL always increases with bending curvature due to the high-order modes have larger loss than low-order modes. This observation has been confirmed by experiments. Finally, we compare the multi-segment FMFs model with and without considering the accompanied MDL, the results show that micro-bending would degrade the system performance especially when considering the accompanied MDL.

关键词：

Realization of phase locking in good-bad-cavity active optical clock

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Shi, Tiantian Pan, Duo Chen, Jingbiao State Key Laboratory of Advanced Optical Communication System and Network School of Electronics Engineering and Computer Science Peking University Beijing100871 China

The residual cavity-pulling effect limits further narrowing of linewidth in dual-wavelength (DW) good-bad-cavity active optical clocks (AOCs). In this paper, we for the first time experimentally realize the cavity-length stabilization of the 1064/1470 nm DW-AOCs by utilizing the phase locking technique of two independent 1064 nm good-cavity lasers. The frequency tracking accuracy between the two main-cavities of DW-AOCs is better than 3 × 10−16 at 1 s, and can reach 1 × 10−17 at 1000 s. Each 1470 nm bad-cavity laser achieves a most probable linewidth of 53 Hz, which is about a quarter of that without phase locking. The influence of the asynchronous cavity-lengths variation between two DW laser systems is suppressed. Copyright © 2019, The Authors. All rights reserved.

关键词： Atomic clocks

Architecture and Performance of a Bidirectional OXC Based on Reversible optical Switches with Reduced Complexity

学校读者我要写书评

暂无评论

Architecture and Performance of a Bidirectional OXC Based on...

33rd European Conference and Exhibition of optical communication

作者： Yongmin Qi Weisheng Hu Yi Dong Wen-De Zhong Xinyu Xu Hao He Yaohui Jin Wei Guo Weiqiang Sun Shilin Xiao Yikai Su State Key Lab of Advanced Optical Communication System and Network Shanghai Jiao Tong University Shanghai 200240 China Network Technology Research Centre Nanyang Technological University Singapore 639798

By exploiting the symmetry of bidirectional wavelength-connections in WDM networks, we propose an NN bidirectional OXC using one N/2N/2 reversible optical switch to reduce the complexity of the OXCs. The feasibility i... 详细信息

关键词： optical reflection optical modulation Micromechanical devices

Dependent multi-task learning with causal intervention for image captioning

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Chen, Wenqing Tian, Jidong Fan, Caoyun He, Hao Jin, Yaohui MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University China State Key Lab of Advanced Optical Communication System and Network Shanghai Jiao Tong University China

Recent work for image captioning mainly followed an extract-then-generate paradigm, pre-extracting a sequence of object-based features and then formulating image captioning as a single sequence-to-sequence task. Although promising, we observed two problems in generated captions: 1) content inconsistency where models would generate contradicting facts;2) not informative enough where models would miss parts of important information. From a causal perspective, the reason is that models have captured spurious statistical correlations between visual features and certain expressions (e.g., visual features of "long hair" and "woman"). In this paper, we propose a dependent multi-task learning framework with the causal intervention (DMTCI). Firstly, we involve an intermediate task, bag-of-categories generation, before the final task, image captioning. The intermediate task would help the model better understand the visual features and thus alleviate the content inconsistency problem. Secondly, we apply Pearl’s do-calculus on the model, cutting off the link between the visual features and possible confounders and thus letting models focus on the causal visual features. Specifically, the high-frequency concept set is considered as the proxy confounders where the real confounders are inferred in the continuous space. Finally, we use a multi-agent reinforcement learning (MARL) strategy to enable end-to-end training and reduce the inter-task error accumulations. The extensive experiments show that our model outperforms the baseline models and achieves competitive performance with state-of-the-art models. © 2021, CC BY.

关键词： Fertilizers