Is "Deep Learning" Fraudulent in Statistics?
Pages 1 - 8
Abstract
This is the third theoretical paper on “Deep Learning” misconduct, addressing the statistical aspect. The first and second papers on Deep Learning misconduct are [26, 27]. Regardless of learning modes, e.g., supervised, reinforcement, adversarial, and evolutional, almost all Deep Learning projects are rooted in the same misconduct—cheating and hiding—cheating by reporting fit error as test error and hiding bad data. This paper presents new mathematical results that explain why Deep Learning is fraudulent in statistics. Furthermore, this paper presents new statistical reasons why authors must report at least the average error of all trained networks, good and bad, on the validation set along with their standard deviation. For the first time, this paper reveals that both PSUTS (Post-Selection Using Test Set) and PSUVS (Post-Selection Using Validation Set) egregiously replace the mean of random samples with the smallest sample. The paper further alleges that more recent Deep Learning systems such as Transformer, ChatGPT, Bard, and AlphaDev are also fraudulent because they are based on the same Deep Learning fraud. Detailed evidence of the alleged frauds is beyond the scope of this paper and should be heard by a court.
Supplemental Material
ZIP File
Zipped directory with latex source files.
- Download
- 7.18 MB
References
[1]
R. Berk, L. Brown, A. Buja, K. Zhang, and L. Zhao. 2013. Valid Post-Selection Inference. The Annals of Statistics 41, 2 (2013), 802–837.
[2]
V. Cherkassky and F. Mulier. 1998. Learning from Data. Wiley, New York.
[3]
R. O. Duda, P. E. Hart, and D. G. Stork. 2001. Pattern Classification (2nd ed.). Wiley, New York.
[4]
A. A. Giordano and F. M. Hsu. 1985. Least Square Estimation with Applications to Digital Signal Processing. John Wiley & Sons, New York.
[5]
A. Graves, G. Wayne, M. Reynolds, D. Hassabis, 2016. Hybrid computing using a neural network with dynamic external memory. Nature 538 (2016), 471–476.
[6]
S. Hochreiter and J. Schmidhuber. 1997. Long Short-Term Memory. Neural Computation 9, 8 (1997), 1735–1780.
[7]
G. B. Huang and C. K.Siew. 2006. Universal Approximation Using Incremental Constructive Feedforward Networks With Random Hidden Nodes. IEEE Transactions on Neural Networks 17, 4 (2006), 879–892.
[8]
G. B. Huang, K.Z. Mao, C. K. Siew, and D. S. Huang. 2005. Fast Modular network implementation for support vector machines. IEEE Transactions on Neural Networks 16, 6 (2005), 1651– 1663.
[9]
A. Krizhevsky, I. Sutskever, and G. E. Hinton. 2017. Imagenet classification with deep convolutional neural networks. Commun. ACM 60, 6 (2017), 84–90.
[10]
D. J. Mankowitz, D. J. Michi, D. Silver, 2023. Faster sorting algorithms discovered using deep reinforcement learning. Nature 618 (2023), 257–263.
[11]
G. J. McLachlan. 1992. Discriminant Analysis and Statistical Pattern Recognition. Wiley, New York.
[12]
O. Russakovsky, J. Deng, L. Fei-Fei, 2015. ImageNet Large Scale Visual Recognition Challenge. Int’l Journal of Computer Vision 115 (2015), 211–252.
[13]
V. Saggio, B. E. Asenbeck, P. Walther, 2021. Experimental quantum speed-up in reinforcement learning agents. Nature 591, 7849 (March 11 2021), 229–233.
[14]
J. Schrittwieser, I. Antonoglou, D. Silver, 2020. Mastering Atari, Go, chess and shogi by planning with a learned model. Science 588, 7839 (2020), 604–609.
[15]
A. W. Senior, R. Evans, D. Hassabis, 2020. Improved protein structure prediction using potentials from deep learning. Nature 577 (2020), 706–710.
[16]
D. Silver, A. Huang, D. Hassabis, 2016. Mastering the game of Go with deep neural networks and tree search. Nature 529 (January 27 2016), 484–489.
[17]
D. Silver, T. Hubert, D. Hassabis, 2018. A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play. Science 362, 6419 (2018), 1140–1144.
[18]
D. Silver, J. Schrittwieser, D. Hassabis, 2017. Mastering the game of Go without human knowledge. Nature 550 (2017), 354–359.
[19]
N. Slonim, Y. Bilu, C. Alzate, R. Aharonov, 2021. An autonomous debating system. Nature 591, 7850 (March 18 2021), 379–384.
[20]
A. Vaswani, N. Shazeer, I. Polosukhin, 2017. Attemtion Is All You Need. In Proc. Neural Info. Proc. Systems (NIPS). NIPS Foundation, Long Beach, CA, 1–15.
[21]
J. Weng. 2011. Why Have We Passed “neural networks do not abstract well”?Natural Intelligence: the INNS Magazine 1, 1 (2011), 13–22.
[22]
J. Weng. 2015. Brain as an Emergent Finite Automaton: A Theory and Three Theorems. Int’l Journal of Intelligence Science 5, 2 (2015), 112–131.
[23]
J. Weng. 2021. Data Deletions in AI Papers in Nature since 2015 and the Appropriate Protocol. http://www.cse.msu.edu/ weng/research/2021-06-28-Report-to-Nature-specific-PSUTS.pdf. submitted to Nature, June 28, 2021.
[24]
J. Weng. 2021. Data Deletions in AI Papers in Science since 2015 and the Appropriate Protocol. http://www.cse.msu.edu/ weng/research/2021-12-13-Report-to-Science-specific-PSUTS.pdf. submitted to Science, Dec. 13, 2021.
[25]
J. Weng. 2022. 20 million-dollar problems for any brain models and a holistic solution: Conscious learning. In Proc. Int’l Joint Conference on Neural Networks. NJ: IEEE Press, Padua, Italy, 1–9. http://www.cse.msu.edu/ weng/research/20M-IJCNN2022rvsd-cite.pdf.
[26]
J. Weng. 2022. 3D-to-2D-to-3D Conscious Learning. In Proc. IEEE 40th Int’l Conference on Consumer Electronics. NJ: IEEE Press, Las Vegas, NV, USA, 1–6.
[27]
J. Weng. 2022. An Algorithmic Theory of Conscious Learning. In 2022 3rd Int’l Conf. on Artificial Intelligence in Electronics Engineering. NY: ACM Press, Bangkok, Thailand, 1–10.
[28]
J. Weng. 2022. A Developmental Network Model of Conscious Learning in Biological Brains. Research Square., 32 pages.
[29]
J. Weng. 2022. On “Deep Learning” Misconduct. In Proc. 2022 3rd International Symposium on Automation, Information and Computing (ISAIC 2022). SciTePress, Beijing, China, 1–8. arXiv:2211.16350.
[30]
J. Weng. 2023. A Protocol for Testing Conscious Learning Robots. In Proc. Int’l Joint Conference on Neural Networks. NJ: IEEE Press, Queensland, Australia, 1–8.
[31]
J. Weng. 2023. Why Deep Learning’s Performance Data Are Misleading. In 2023 4th Int’l Conf. on Artificial Intelligence in Electronics Engineering. NY: ACM Press, Haikou, China, 1–10. arXiv:2208.11228.
[32]
J. Weng, N. Ahuja, and T. S. Huang. 1997. Learning recognition and segmentation using the Cresceptron. Int’l Journal of Computer Vision 25, 2 (Nov. 1997), 109–143.
Index Terms
- Is "Deep Learning" Fraudulent in Statistics?
Recommendations
Deep Reinforcement Learning: From Q-Learning to Deep Q-Learning
Neural Information ProcessingAbstractAs the two hottest branches of machine learning, deep learning and reinforcement learning both play a vital role in the field of artificial intelligence. Combining deep learning with reinforcement learning, deep reinforcement learning is a method ...
Comments
Information & Contributors
Information
Published In

January 2024
89 pages
ISBN:9798400716850
DOI:10.1145/3658835
Copyright © 2024 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 12 June 2024
Check for updates
Qualifiers
- Research-article
- Research
- Refereed limited
Data Availability
Zipped directory with latex source files. https://dl.acm.org/doi/10.1145/3658835.3658836#aiee24-3002.zip
Conference
AIEE 2024
AIEE 2024: 2024 5th International Conference on Artificial Intelligence in Electronics Engineering
January 15 - 17, 2024
Bangkok, Thailand
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 31Total Downloads
- Downloads (Last 12 months)31
- Downloads (Last 6 weeks)5
Reflects downloads up to 16 Feb 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign inFull Access
View options
View or Download as a PDF file.
PDFeReader
View online with eReader.
eReaderHTML Format
View this article in HTML Format.
HTML Format