Skip to main content

Advertisement

Springer Nature Link
Log in

Enhancing Pre-trained Deep Learning Model with Self-Adaptive Reflection

  • Research
  • Published:
Cognitive Computation Aims and scope Submit manuscript

Abstract

In the text mining area, prevalent deep learning models primarily focus on mapping input features to result of predicted outputs, which exhibit a deficiency in self-dialectical thinking process. Inspired by self-reflective mechanisms in human cognition, we propose a hypothesis that existing models emulate decision-making processes and automatically rectify erroneous predictions. The Self-adaptive Reflection Enhanced pre-trained deep learning Model (S-REM) is introduced to validate our hypotheses and to determine the types of knowledge that warrant reproduction. Based on the pretrained-model, S-REM introduces the local explanation for pseudo-label and the global explanation for all labels as the explanation knowledge. The keyword knowledge from TF-IDF model is also integrated to form a reflection knowledge. Based on the key explanation features, the pretrained-model reflects on the initial decision by two reflection methods and optimizes the prediction of deep learning models. Experiments with local and global reflection variants of S-REM on two text mining tasks across four datasets, encompassing three public and one private dataset were conducted. The outcomes demonstrate the efficacy of our method in improving the accuracy of state-of-the-art deep learning models. Furthermore, the method can serve as a foundational step towards developing explainable through integration with various deep learning models.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Explore related subjects

Discover the latest articles and news from researchers in related subjects, suggested using machine learning.

Data Availability

The data and code can be shared via email once the work is published.

References

  1. Bellagente M, Brack M, Teufel H, et al. Multifusion: fusing pre-trained models for multi-lingual, multi-modal image generation. Adv Neural Inf Process Syst. 2024;36.

  2. Dai R. Text Data mining algorithm combining CNN and DBM models. Mob Inf Syst. 2021;2021:1–7.

    Google Scholar 

  3. Sajda P, Philiastides MG, Parra LC. Single-trial analysis of neuroimaging data: inferring neural networks underlying perceptual decision-making in the human brain. IEEE Rev Biomed Eng. 2009;2:97–109.

    Article  Google Scholar 

  4. Akhtar N, Jalwana MAAK. Towards credible visual model interpretation with path attribution[C]//International Conference on Machine Learning. PMLR. 2023;439–457.

  5. Lewis PR, Sarkadi Ş. Reflective artificial intelligence. Mind Mach. 2024;34(2):1–30.

    Article  Google Scholar 

  6. Campbell GE, Bolton AE. Fitting human data with fast, frugal, and computable models of decision-making. InProceedings of the Human Factors and Ergonomics Society Annual Meeting 2003 Oct (Vol. 47, No. 3, pp. 325–329). Sage CA: Los Angeles, CA: SAGE Publications.

  7. Kim B, Park J, Suh J. Transparency and accountability in AI decision support: explaining and visualizing convolutional neural networks for text information. Decis Support Syst. 2020;134:113302.

    Article  Google Scholar 

  8. Cao M, Stewart A, Leonard NE. Integrating human and robot decision-making dynamics with feedback: models and convergence analysis. In2008 47th IEEE Conference on Decision and Control. IEEE. 2008;1127–1132.

  9. Hu Z, Shao M, Liu H, Mi J. Cognitive computing and rule extraction in generalized one-sided formal contexts. Cogn Comput. 2022;14(6):2087–107.

    Article  Google Scholar 

  10. Zuo G, Pan T, Zhang T, Yang Y. SOAR improved artificial neural network for multistep decision-making tasks. Cogn Comput. 2021;13:612–25.

    Article  Google Scholar 

  11. Young T, Hazarika D, Poria S, Cambria E. Recent trends in deep learning based natural language processing. IEEE Comput Intell Mag. 2018;13(3):55–75.

    Article  Google Scholar 

  12. Hilzensauer W. Theoretische Zugänge und Methoden zur Reflexion des Lernens. Ein Diskussionsbeitrag. Bildungsforschung. 2008;2.

  13. Leary MR. The curse of the self: self-awareness, egotism, and the quality of human life. Oxford University Press; 2007.

  14. Ribeiro MT, Singh S, Guestrin C. “Why should I trust you?” Explaining the predictions of any classifier. InProceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 2016;1135–1144.

  15. Wang Q, Mao Z, Wang B, Guo L. Knowledge graph embedding: a survey of approaches and applications. IEEE Trans Knowl Data Eng. 2017;29(12):2724–43.

    Article  Google Scholar 

  16. Dettmers T, Minervini P, Stenetorp P, Riedel S. Convolutional 2d knowledge graph embeddings. Proc AAAI Conf Artif Intell. 2018;32(1).

  17. Quinn CJ, Kiyavash N, Coleman TP. Directed information graphs. IEEE Trans Inf Theory. 2015;61(12):6887–909.

    Article  MathSciNet  Google Scholar 

  18. Weiss K, Khoshgoftaar TM, Wang D. A survey of transfer learning. J Big Data. 2016;3(1):1–40.

    Article  Google Scholar 

  19. Pan SJ, Yang Q. A survey on transfer learning. IEEE Trans Knowl Data Eng. 2009;22(10):1345–59.

    Article  Google Scholar 

  20. Nguyen BH, Xue B, Andreae P, Zhang M. A hybrid evolutionary computation approach to inducing transfer classifiers for domain adaptation. IEEE Trans Cybern. 2020;51(12):6319–32.

    Article  Google Scholar 

  21. Zhao H, Sun X, Dong J, Chen C, Dong Z. Highlight every step: knowledge distillation via collaborative teaching. IEEE Trans Cybern. 2020;52(4):2070–81.

    Article  Google Scholar 

  22. Zhang J, Chen B, Zhang L, Ke X, Ding H. Neural, symbolic and neural-symbolic reasoning on knowledge graphs. AI Open. 2021;2:14–35.

    Article  Google Scholar 

  23. Hooker JN. A quantitative approach to logical inference. Decis Support Syst. 1988;4(1):45–69.

    Article  Google Scholar 

  24. Deng H. Interpreting tree ensembles with inTrees. Int J Data Sci Anal. 2019;7(4):277–87.

    Article  Google Scholar 

  25. Mashayekhi M, Gras R. Rule extraction from random forest: the RF+ HC methods. InAdvances in Artificial Intelligence: 28th Canadian Conference on Artificial Intelligence, Canadian AI 2015, Halifax, Nova Scotia, Canada, June 2–5, 2015, Proceedings 28 2015 (pp. 223–237). Springer International Publishing. https://doi.org/10.1007/978-3-319-18356-5_20.

  26. Puri N, Gupta P, Agarwal P, Verma S, Krishnamurthy B. Magix: Model agnostic globally interpretable explanations. arXiv preprint arXiv:1706.07160. 2017 Jun 22. https://doi.org/10.48550/arXiv.1706.07160.

  27. Yang C, Rangarajan A, Ranka S. Global model interpretation via recursive partitioning. In2018 IEEE 20th International Conference on High Performance Computing and Communications; IEEE 16th International Conference on Smart City; IEEE 4th International Conference on Data Science and Systems (HPCC/SmartCity/DSS). IEEE. 2018;1563–1570. https://doi.org/10.1109/HPCC/SmartCity/DSS.2018.00256.

  28. Yuan H, Chen Y, Hu X, Ji S. Interpreting deep models for text analysis via optimization and regularization methods. Proc AAAI Conf Artif Intell. 2019;33(01):5717–24.

    Google Scholar 

  29. Mahendran A, Vedaldi A. Understanding deep image representations by inverting them. Proc IEEE Conf Comput Vision Pattern Recogn. 2015;5188–5196.

  30. Dosovitskiy A, Brox T. Inverting visual representations with convolutional networks. InProceedings of the IEEE conference on computer vision and pattern recognition. 2016;4829–4837.

  31. Guidotti R, Monreale A, Ruggieri S, Pedreschi D, Turini F, Giannotti F. Local rule-based explanations of black box decision systems. arXiv preprint arXiv:1805.10820. 2018 May 28. https://doi.org/10.48550/arXiv.1805.10820.

  32. Ribeiro MT, Singh S, Guestrin C. Anchors: High-precision model-agnostic explanations. Proc AAAI Conf Artif Intell. 2018;32(1). https://doi.org/10.1609/aaai.v32i1.11491.

  33. Liu L, Wang L. What has my classifier learned? visualizing the classification rules of bag-of-feature model by support region detection. 2012 IEEE Conf Comput Vision Pattern Recogn IEEE. 2012;3586–3593.

  34. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: visual explanations from deep networks via gradient-based localization. Proc IEEE Int Conf Comput Vision. 2017;618–626.

  35. Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 2017;30.

  36. Guo W, Mu D, Xu J, Su P, Wang G, Xing X. Lemna: explaining deep learning based security applications. Inproceedings of the 2018 ACM SIGSAC conference on computer and communications security. 2018;364–379.

  37. Li X, Xiong H, Li X, et al. G-LIME: statistical learning for local interpretations of deep neural networks using global priors. Artif Intell. 2023;314:103823.

    Article  MathSciNet  Google Scholar 

  38. Chiu CW, Minku LL. A diversity framework for dealing with multiple types of concept drift based on clustering in the model space. IEEE Trans Neural Netw Learn Syst. 2020;33(3):1299–309.

    Article  Google Scholar 

  39. Liu S, Xue S, Wu J, Zhou C, Yang J, Li Z, Cao J. Online active learning for drifting data streams. IEEE Trans Neural Netw Learn Syst. 2021. https://doi.org/10.1109/TNNLS.2021.3091681.

    Article  Google Scholar 

  40. Bi X, Zhang C, Zhao X, Li D, Sun Y, Ma Y. CODES: Efficient incremental semi-supervised classification over drifting and evolving social streams. IEEE Access. 2020;8:14024–35. https://doi.org/10.1109/ACCESS.2020.2965766.

    Article  Google Scholar 

  41. Li H, Dong W, Hu BG. Incremental concept learning via online generative memory recall. IEEE Trans Neural Netw Learn Syst. 2020;32(7):3206–16. https://doi.org/10.1109/TNNLS.2020.3010581.

    Article  Google Scholar 

  42. Shan J, Zhang H, Liu W, Liu Q. Online active learning ensemble framework for drifted data streams. IEEE Trans Neural Netw Learn Syst. 2018;30(2):486–98.

    Article  Google Scholar 

  43. Petit G, Popescu A, Schindler H, et al. Fetril: feature translation for exemplar-free class-incremental learning[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2023;3911–3920.

  44. Li P, He L, Wang H, Hu X, Zhang Y, Li L, Wu X. Learning from short text streams with topic drifts. IEEE Trans Cybern. 2017;48(9):2697–711.

    Article  Google Scholar 

  45. Lu Y, Cheung YM, Tang YY. Adaptive chunk-based dynamic weighted majority for imbalanced data streams with concept drift. IEEE Trans Neural Netw Learn Syst. 2019;31(8):2764–78.

    Article  Google Scholar 

  46. Yang C, Cheung YM, Ding J, Tan KC. Concept drift-tolerant transfer learning in dynamic environments. IEEE Trans Neural Netw Learn Syst. 2021;33(8):3857–71.

    Article  Google Scholar 

  47. Pan Z, Yu X, Zhang M, et al. DyCR: a dynamic clustering and recovering network for few-shot class-incremental learning. IEEE Trans Neural Netw Learn Syst. 2024.

  48. Gehring J, Auli M, Grangier D, Dauphin YN. A convolutional encoder model for neural machine translation. arXiv preprint arXiv:1611.02344. 2016 Nov 7. https://doi.org/10.48550/arXiv.1611.02344.

  49. Bartoli A, De Lorenzo A, Medvet E, Tarlao F. Active learning of regular expressions for entity extraction. IEEE Trans Cybern. 2017;48(3):1067–80.

    Article  Google Scholar 

  50. Jiang H, He H. Learning from negative links. IEEE Trans Cybern. 2021;52(8):8481–92.

    Article  Google Scholar 

  51. Wu Y, Dong Y, Qin J, Pedrycz W. Linguistic distribution and priority-based approximation to linguistic preference relations with flexible linguistic expressions in decision making. IEEE Trans Cybern. 2020;51(2):649–59.

    Article  Google Scholar 

  52. Pang J, Rao Y, Xie H, Wang X, Wang FL, Wong TL, Li Q. Fast supervised topic models for short text emotion detection. IEEE Trans Cybern. 2019;51(2):815–28.

    Article  Google Scholar 

  53. Wang X, Kou L, Sugumaran V, Luo X, Zhang H. Emotion correlation mining through deep learning models on natural language text. IEEE Trans Cybern. 2021;51(9):4400–13.

  54. Wu Z, Ong DC. Context-guided bert for targeted aspect-based sentiment analysis. Proc AAAI Conf Artif Intell. 2021;35(16):14094–102.

    Google Scholar 

  55. Wu HC, Luk RW, Wong KF, Kwok KL. Interpreting TF-IDF term weights as making relevance decisions. ACM Trans Inf Syst (TOIS). 2008;26(3):1–37.

    Article  Google Scholar 

  56. Pontiki M, Galanis D, Papageorgiou H, Androutsopoulos I, Manandhar S, AL-Smadi M, Al-Ayyoub M, Zhao Y, Qin B, De Clercq O, Hoste V. Semeval-2016 task 5: aspect based sentiment analysis. InProWorkshop on Semantic Evaluation (SemEval-2016). Assoc Comput Linguist. 2016;19–30.

  57. Zhang X, Zhao J, LeCun Y. Character-level convolutional networks for text classification. Adv Neural Inf Process Syst. 2015;28.

Download references

Funding

The research reported in this paper was supported by the National Natural Science Foundation of China under the grant 72204155 and Natural Science Foundation of Shanghai under the grant 23ZR1423100.

Author information

Authors and Affiliations

  1. School of Computer Engineering and Science, Shanghai University, Shanghai, 200444, China

    Xinzhi Wang, Mengyue Li & Hang Yu

  2. School of Safety Science, Tsinghua University, Beijing, 100084, China

    Chenyang Wang & Hui Zhang

  3. Department of Decision and Information Sciences, Oakland University, Rochester, MI, 48309, USA

    Vijayan Sugumaran

Authors
  1. Xinzhi Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Mengyue Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Hang Yu
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Chenyang Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Vijayan Sugumaran
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Hui Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Methodology and resources were provided by Xinzhi Wang. Review, editing, and original draft preparation were done by Xinzhi Wang and Mengyue Li. Investigation and visualization were done by Mengyue Li and Chenyang Wang. Software and data curation were done by Mengyue Li. Conceptualization and formal analysis were discussed by Hang Yu and Vijayan Sugumaran. This work was supervised by Hang Yu and Xinzhi Wang.

Corresponding authors

Correspondence to Hang Yu or Chenyang Wang.

Ethics declarations

Competing Interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, X., Li, M., Yu, H. et al. Enhancing Pre-trained Deep Learning Model with Self-Adaptive Reflection. Cogn Comput 16, 3468–3483 (2024). https://doi.org/10.1007/s12559-024-10348-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12559-024-10348-3

Keywords

Profiles

  1. Xinzhi Wang View author profile