Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Time-Aware Knowledge Representations of Dynamic Objects with Multidimensional Persistence (2401.13157v1)

Published 24 Jan 2024 in cs.LG and cs.AI

Abstract: Learning time-evolving objects such as multivariate time series and dynamic networks requires the development of novel knowledge representation mechanisms and neural network architectures, which allow for capturing implicit time-dependent information contained in the data. Such information is typically not directly observed but plays a key role in the learning task performance. In turn, lack of time dimension in knowledge encoding mechanisms for time-dependent data leads to frequent model updates, poor learning performance, and, as a result, subpar decision-making. Here we propose a new approach to a time-aware knowledge representation mechanism that notably focuses on implicit time-dependent topological information along multiple geometric dimensions. In particular, we propose a new approach, named \textit{Temporal MultiPersistence} (TMP), which produces multidimensional topological fingerprints of the data by using the existing single parameter topological summaries. The main idea behind TMP is to merge the two newest directions in topological representation learning, that is, multi-persistence which simultaneously describes data shape evolution along multiple key parameters, and zigzag persistence to enable us to extract the most salient data shape information over time. We derive theoretical guarantees of TMP vectorizations and show its utility, in application to forecasting on benchmark traffic flow, Ethereum blockchain, and electrocardiogram datasets, demonstrating the competitive performance, especially, in scenarios of limited data records. In addition, our TMP method improves the computational efficiency of the state-of-the-art multipersistence summaries up to 59.5 times.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (56)
  1. Adams, H.; et al. 2017. Persistence images: A stable vector representation of persistent homology. JMLR, 18.
  2. Blockchain networks: Data structures of bitcoin, monero, zcash, ethereum, ripple, and iota. Wiley Int. Reviews: Data Mining and Knowledge Discovery, 12(1): e1436.
  3. Persistence homology of networks: methods and applications. Applied Network Science, 4(1): 1–28.
  4. Ali, D.; et al. 2023. A survey of vectorization methods in topological data analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence.
  5. Atienza, N.; et al. 2020. On the stability of persistent entropy and new summary functions for topological data analysis. Pattern Recognition, 107: 107509.
  6. Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting. NeurIPS, 33.
  7. Bin, Y.; et al. 2018. Describing video with attention-based bidirectional LSTM. IEEE transactions on cybernetics, 49(7): 2631–2641.
  8. An introduction to multiparameter persistence. arXiv preprint arXiv:2203.14289.
  9. Signed barcodes for multi-parameter persistence via rank decompositions. In SoCG.
  10. Bubenik, P. 2015. Statistical Topological Data Analysis using Persistence Landscapes. JMLR, 16(1): 77–102.
  11. Cao, D.; et al. 2020. Spectral Temporal Graph Neural Network for Multivariate Time-series Forecasting. In NeurIPS, volume 33, 17766–17778.
  12. Zigzag persistent homology and real-valued functions. In SoCG.
  13. Zigzag Persistence. Found. Comput. Math., 10(4): 367–405.
  14. Multiparameter persistence image for topological machine learning. NeurIPS.
  15. Stochastic convergence of persistence landscapes and silhouettes. In SoCG.
  16. Freeway performance measurement system: mining loop detector data. Transportation Research Record, 1748(1): 96–102.
  17. The UCR time series classification archive.
  18. Z-GCNETs: time zigzags at graph convolutional networks for time series forecasting. In ICML, 1684–1694. PMLR.
  19. TAMP-S2GCNets: coupling time-aware multipersistence knowledge representation with spatio-supra graph convolutional networks for time-series forecasting. In ICLR.
  20. Chen, Y.; et al. 2015b. A general framework for never-ending learning from time series streams. Data mining and knowledge discovery, 29: 1622–1664.
  21. Persistence curves: A canonical framework for summarizing persistence diagrams. Advances in Computational Mathematics, 48(1): 6.
  22. A kernel for multi-parameter persistent homology. Computers & graphics: X, 2: 100005.
  23. Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering. In NeurIPS, volume 29, 3844–3852.
  24. Gate-variants of Gated Recurrent Unit (GRU) neural networks. In MWSCAS.
  25. Computational Topology for Data Analysis. Cambridge University Press.
  26. Tokens, Types, and Standards: Identification and Utilization in Ethereum. In 2020 IEEE DAPPS, 1–10.
  27. Computational topology: an introduction. American Mathematical Soc.
  28. Eisenbud, D. 2013. Commutative algebra: with a view toward algebraic geometry, volume 150. Springer Science & Business Media.
  29. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In AAAI.
  30. Graph filtration learning. In ICML, 4314–4323. PMLR.
  31. Graph neural network for traffic forecasting: A survey. Expert Systems with Applications, 207: 117921.
  32. Instability of the Betti Sequence for Persistent Homology. J. Korean Soc. Ind. and Applied Math., 25(4): 296–311.
  33. Semi-supervised classification with graph convolutional networks. ICLR.
  34. A survey on graph kernels. Applied Network Science, 1–42.
  35. Computing minimal presentations and bigraded betti numbers of 2-parameter persistent homology. SIAGA.
  36. Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting. In ICLR.
  37. A critical review of recurrent neural networks for sequence learning. arXiv preprint arXiv:1506.00019.
  38. Evolvegcn: Evolving graph convolutional networks for dynamic graphs. In AAAI.
  39. Schmidhuber, J. 2017. LSTM: Impact on the world’s most valuable public companies. http://people.idsia.ch/~juergen/impact-on-most-valuable-companies.html. Accessed: 2020-03-19.
  40. Segovia Dominguez, I.; et al. 2021. Does Air Quality Really Impact COVID-19 Clinical Severity: Coupling NASA Satellite Datasets with Geometric Deep Learning. KDD.
  41. Segovia-Dominguez, I.; et al. 2021. TLife-LSTM: Forecasting Future COVID-19 Progression with Topological Signatures of Atmospheric Conditions. In PAKDD (1), 201–212.
  42. Chartalist: Labeled Graph Datasets for UTXO and Account-based Blockchains. NeurIPS.
  43. Performance evaluation of deep neural networks applied to speech recognition: RNN, LSTM and GRU. J. Artificial Intelligence and Soft Computing Research, 9(4): 235–245.
  44. Skeleton-Based Dynamic Hand Gesture Recognition Using a Part-Based GRU-RNN for Gesture-Based Interface. IEEE Access, 8: 50236–50243.
  45. Sequence to Sequence Learning with Neural Networks. In Proceedings of the 27th International Conference on Neural Information Processing Systems - Volume 2, NIPS’14, 3104–3112. Cambridge, MA, USA: MIT Press.
  46. Finding the Smallest H-Subgraph in Real Weighted Graphs and Related Problems. In Automata, Languages and Programming.
  47. Graph attention networks. ICLR.
  48. Vipond, O. 2020. Multiparameter Persistence Landscapes. J. Mach. Learn. Res., 21: 61–1.
  49. Image Captioning with Deep Bidirectional LSTMs and Multi-Task Learning. ACM Trans. Multimedia Comput. Commun. Appl., 14(2s).
  50. Weber, M.; et al. 2019. Anti-Money Laundering in Bitcoin: Experimenting with GCNs for Financial Forensics. In KDD.
  51. Graph WaveNet for Deep Spatial-Temporal Graph Modeling. In IJCAI.
  52. A rainfall-runoff model with LSTM-based sequence-to-sequence learning. Water resources research, 56(1): e2019WR025326.
  53. Spatial temporal graph convolutional networks for skeleton-based action recognition. In AAAI.
  54. Deep multi-view spatial-temporal network for taxi demand prediction. In AAAI.
  55. Spatio-temporal GCNs: a deep learning framework for traffic forecasting. In IJCAI.
  56. A Review of Recurrent Neural Networks: Lstm Cells and Network Architectures. Neural Comput., 31(7): 1235–1270.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now