Deep Learning by a Unitary Tensor Network Algorithm Provides Hyperfast Financial Literacy
39 Pages Posted: 6 Feb 2020 Last revised: 24 Jun 2023
Date Written: January 12, 2020
We show how tensor network theory (Orús, 2014) and deep learning can be combined to provide a neural tensor network of financial information for hyperfast financial literacy. The resulting minimal-complexity, 5-layered structure encodes an infinite number of probable outcomes into a graphical alphabet made up by 12 superpositioned binary units called double-entries (see fig 1). Using the proposed financial wave function (Schrödinger, 1935), as a computational resource (Biamonte, 2016), we obtain hyperfast processing of financial statements, one pixel at a time. This reveals a highly entangled architecture (Levine, et al., 2019). Here, complexity scales linearly, not exponentially (Huggins, et al., 2018). This enables quantum states across a phase transition to require only a very small training data set (Caro, M., et al., 2022). With the new algorithm, people can learn financial accounting in 10 hours; a process that would take at least one year with the traditional financial paradigm, observed by Luca Pacioli in 1494. Results are based on solid empirical evidence.
Keywords: Tensor Networks, Deep Learning, Computational Complexity, Algorithms, Emergence, Information Reuse, Financial Literacy, Accounting
Suggested Citation: Suggested Citation