Opening the AI Black Box: Distilling Machine-Learned Algorithms into Code
Name
entropy-26-01046.pdf
Size
685.61 KB
Format
Adobe PDF
Checksum (MD5)
393a123d780b127d8b85f845dd363550
Author(s)
Michaud, Eric J.
Liao, Isaac
Lad, Vedang
Liu, Ziming
Mudide, Anish
Loughridge, Chloe
Guo, Zifan Carl
Kheirkhah, Tara Rezaei
Vukelić, Mateja
Tegmark, Max
Date Issued
December 2, 2024
Journal
Entropy
Publisher
Multidisciplinary Digital Publishing Institute
Citation
Michaud, E.J.; Liao, I.; Lad, V.; Liu, Z.; Mudide, A.; Loughridge, C.; Guo, Z.C.; Kheirkhah, T.R.; Vukelić, M.; Tegmark, M. Opening the AI Black Box: Distilling Machine-Learned Algorithms into Code. Entropy 2024, 26, 1046.
Version
Final published version
Abstract
Can we turn AI black boxes into code? Although this mission sounds extremely challenging, we show that it is not entirely impossible by presenting a proof-of-concept method, MIPS, that can synthesize programs based on the automated mechanistic interpretability of neural networks trained to perform the desired task, auto-distilling the learned algorithm into Python code. We test MIPS on a benchmark of 62 algorithmic tasks that can be learned by an RNN and find it highly complementary to GPT-4: MIPS solves 32 of them, including 13 that are not solved by GPT-4 (which also solves 30). MIPS uses an integer autoencoder to convert the RNN into a finite state machine, then applies Boolean or integer symbolic regression to capture the learned algorithm. As opposed to large language models, this program synthesis technique makes no use of (and is therefore not limited by) human training data such as algorithms and code from GitHub. We discuss opportunities and challenges for scaling up this approach to make machine-learned models more interpretable and trustworthy.
MIT Department
Massachusetts Institute of Technology. Department of Physics
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.3390/e26121046
