ai-descartes's Introduction

AI Descartes: Combining Data and Theory for Derivable Scientific Discovery

This repository contains the code and the data used for the experiments in the paper Combining data and theory for derivable scientific discovery with AI-Descartes.

Visit our website for a general overview, references, and some introductory videos: → AI-Descartes website ←

Folders description:

data: contains the 3 datasets used in the paper (Kepler’s third law of planetary motion, Einstein’s time-dilation formula, Langmuir’s adsorption equation), the data points for 81 FSRD problems and the corresponding background theories (see data/README.md).
reasoning: contains the code for the Reasoning module of AI-Descartes (see reasoning/README.md)
symbolic-regression: contains the code for the Symbolic Regression module of AI-Descartes (see symbolic-regression/README.md)

How to cite

@article{AI_Descartes,
	title = {Combining data and theory for derivable scientific discovery with {AI}-{Descartes}},
	volume = {14},
	issn = {2041-1723},
	url = {https://doi.org/10.1038/s41467-023-37236-y},
	doi = {10.1038/s41467-023-37236-y},
	number = {1},
	journal = {Nature Communications},
	author = {Cornelio, Cristina and Dash, Sanjeeb and Austel, Vernon and Josephson, Tyler R. and Goncalves, Joao and Clarkson, Kenneth L. and Megiddo, Nimrod and El Khadir, Bachir and Horesh, Lior},
	month = apr,
	year = {2023},
	pages = {1777},
}

ai-descartes's People

Contributors

Stargazers

Watchers

ai-descartes's Issues

Found: "Functions" at 10:1 Expected: ("ArchiveEntry" | "Lemma" | "Theorem" | "Exercise") Hint: Try ("ArchiveEntry" | "Lemma" | "Theorem" | "Exercise")

add: ArchiveEntry "" and Real abs( Real x ) =

Help needed to run AI-Descartes symbolic regression

I have some problems running the system with the provided datasets. In particular I get the repeated output that jobs failed

~/AI-Descartes/symbolic-regression/run_emf$ java -cp ../emf/src/:../emf/yamlbeans-1.0.jar  emf.Main opts-common.yaml opts-langmuir.yaml datasets/langmuir/sun_et_al/
Current relative path is: /home/gkronber/AI-Descartes/symbolic-regression/run_emf
YAML OPTS from dir: datasets/langmuir/sun_et_al/ [[outputDir: datasets/langmuir/sun_et_al//tmp], [infile: datasets/langmuir/sun_et_al//input.dat], [, opts-common.yaml], [, opts-langmuir.yaml], [, datasets/langmuir/sun_et_al//opts.yaml]]
YAML STRING infile: datasets/langmuir/sun_et_al//input.dat
YAML  {infile=datasets/langmuir/sun_et_al//input.dat}
YAML opts from string infile: datasets/langmuir/sun_et_al//input.dat: {infile=datasets/langmuir/sun_et_al//input.dat}
YAML opts from opts-common.yaml: infile=... multitask=false nonconst_plusminus=false baron_exec=... baronOpts=... inputSampleSize=50 kill_existing_baron_jobs=false max_num_jobs=45 LicName=... printBaronFilesUpfront=true
YAML opts from opts-langmuir.yaml: infile=... multitask=false nonconst_plusminus=false baron_exec=... baronOpts=... inputSampleSize=50 kill_existing_baron_jobs=false max_num_jobs=45 LicName=... printBaronFilesUpfront=true dimensionVars={x=[1, 0], y=[0, 3]} use_dimensional_analysis=false relative_squared_error=false prod_exp_limit=6 add_noise_to_input=false max_num_consts=1 dimensionUnits=[bar, mm] max_tree_depth=3
WARNING: yaml key max_num_consts is being overridden in file datasets/langmuir/sun_et_al//opts.yaml.
YAML opts from datasets/langmuir/sun_et_al//opts.yaml: infile=... multitask=false nonconst_plusminus=false baron_exec=... baronOpts=... inputSampleSize=50 kill_existing_baron_jobs=false max_num_jobs=45 LicName=... printBaronFilesUpfront=true dimensionVars={x=[1, 0], y=[0, 3]} use_dimensional_analysis=false relative_squared_error=false prod_exp_limit=6 add_noise_to_input=false max_num_consts=4 dimensionUnits=[bar, mm] max_tree_depth=3 operators=[*, /, +] max_var_exp=2
UNOPS [*, /, +] [] []
The following yaml items have non-default values:
  max_num_jobs: 45
  multitask: false
  max_num_consts: 4
  prod_exp_limit: 6
  printBaronFilesUpfront: true
  inputSampleSize: 50
 BOP
 [MaxTime: 60;, PrTimeFreq: 1;, EpsA: 1e-4;, EpsR: 1e-3;]

outputDir: datasets/langmuir/sun_et_al//tmp
Operators: [*, /, +]
 0: [0.07] -> 0.695
 1: [0.11] -> 0.752
 2: [0.2] -> 0.797
 3: [0.31] -> 0.825
 4: [0.56] -> 0.86
 5: [0.8] -> 0.882
 6: [1.07] -> 0.904
 7: [1.46] -> 0.923
 8: [3.51] -> 0.976
 9: [6.96] -> 1.212
10: [12.06] -> 1.371
11: [17.26] -> 1.469
12: [27.56] -> 1.535
13: [41.42] -> 1.577
14: [55.2] -> 1.602
15: [68.95] -> 1.619
16: [86.17] -> 1.632
NUM TREES 31 with max depth 3
        ArithNormTree                            Unnormalized (blank if same)
     0   3: P
     1   7: (P+P)
     2  10: (P+P+P)
     3  13: (P/(P+P))
     4  13: (P+P+P+P)
     5  16: (P/(P+P+P))
     6  16: (P+P+P+P+P)
     7  17: ((P+P)/(P+P))
     8  17: (P+(P/(P+P)))
     9  19: (P/(P+P+P+P))
    10  19: (P+P+P+P+P+P)
    11  20: ((P+P+P)/(P+P))
    12  20: ((P+P)/(P+P+P))
    13  20: (P+P+(P/(P+P)))
    14  21: (P+((P+P)/(P+P)))
    15  22: (P+P+P+P+P+P+P)
    16  23: ((P+P+P)/(P+P+P))
    17  23: ((P+P+P+P)/(P+P))
    18  23: ((P+P)/(P+P+P+P))
    19  23: (P+P+P+(P/(P+P)))
    20  24: (P+P+((P+P)/(P+P)))
    21  25: (P+P+P+P+P+P+P+P)
    22  26: ((P+P+P)/(P+P+P+P))
    23  26: ((P+P+P+P)/(P+P+P))
    24  26: (P+P+P+P+(P/(P+P)))
    25  27: (P+P+P+((P+P)/(P+P)))
    26  27: ((P/(P+P))+(P/(P+P)))
    27  29: ((P+P+P+P)/(P+P+P+P))
    28  30: (P+P+P+P+((P+P)/(P+P)))
    29  31: ((P/(P+P))+((P+P)/(P+P)))
    30  35: (((P+P)/(P+P))+((P+P)/(P+P)))

-----------------------

IGNORING inputSampleSize, since too large (50 >= 17).
Printing all jobs upfront - ubound is 0.0
THESE JOBS FAILED: [0:P, 1:(P+P), 2:(P+P+P), 3:(P/(P+P)), 4:(P+P+P+P), 5:(P/(P+P+P)), 6:(P+P+P+P+P), 7:((P+P)/(P+P)), 8:(P+(P/(P+P))), 9:(P/(P+P+P+P)), 10:(P+P+P+P+P+P), 11:((P+P+P)/(P+P)), 12:((P+P)/(P+P+P)), 13:(P+P+(P/(P+P))), 14:(P+((P+P)/(P+P))), 15:(P+P+P+P+P+P+P), 16:((P+P+P)/(P+P+P)), 17:((P+P+P+P)/(P+P)), 18:((P+P)/(P+P+P+P)), 19:(P+P+P+(P/(P+P))), 20:(P+P+((P+P)/(P+P))), 21:(P+P+P+P+P+P+P+P), 22:((P+P+P)/(P+P+P+P)), 23:((P+P+P+P)/(P+P+P)), 24:(P+P+P+P+(P/(P+P))), 25:(P+P+P+((P+P)/(P+P))), 26:((P/(P+P))+(P/(P+P))), 27:((P+P+P+P)/(P+P+P+P)), 28:(P+P+P+P+((P+P)/(P+P))), 29:((P/(P+P))+((P+P)/(P+P))), 30:(((P+P)/(P+P))+((P+P)/(P+P)))]
RERUNNING
.still running: []
THESE JOBS FAILED: [0:P, 1:(P+P), 2:(P+P+P), 3:(P/(P+P)), 4:(P+P+P+P), 5:(P/(P+P+P)), 6:(P+P+P+P+P), 7:((P+P)/(P+P)), 8:(P+(P/(P+P))), 9:(P/(P+P+P+P)), 10:(P+P+P+P+P+P), 11:((P+P+P)/(P+P)), 12:((P+P)/(P+P+P)), 13:(P+P+(P/(P+P))), 14:(P+((P+P)/(P+P))), 15:(P+P+P+P+P+P+P), 16:((P+P+P)/(P+P+P)), 17:((P+P+P+P)/(P+P)), 18:((P+P)/(P+P+P+P)), 19:(P+P+P+(P/(P+P))), 20:(P+P+((P+P)/(P+P))), 21:(P+P+P+P+P+P+P+P), 22:((P+P+P)/(P+P+P+P)), 23:((P+P+P+P)/(P+P+P)), 24:(P+P+P+P+(P/(P+P))), 25:(P+P+P+((P+P)/(P+P))), 26:((P/(P+P))+(P/(P+P))), 27:((P+P+P+P)/(P+P+P+P)), 28:(P+P+P+P+((P+P)/(P+P))), 29:((P/(P+P))+((P+P)/(P+P))), 30:(((P+P)/(P+P))+((P+P)/(P+P)))]
RERUNNING
.THESE JOBS FAILED: [0:P, 1:(P+P), 2:(P+P+P), 3:(P/(P+P)), 4:(P+P+P+P), 5:(P/(P+P+P)), 6:(P+P+P+P+P), 7:((P+P)/(P+P)), 8:(P+(P/(P+P))), 9:(P/(P+P+P+P)), 10:(P+P+P+P+P+P), 11:((P+P+P)/(P+P)), 12:((P+P)/(P+P+P)), 13:(P+P+(P/(P+P))), 14:(P+((P+P)/(P+P))), 15:(P+P+P+P+P+P+P), 16:((P+P+P)/(P+P+P)), 17:((P+P+P+P)/(P+P)), 18:((P+P)/(P+P+P+P)), 19:(P+P+P+(P/(P+P))), 20:(P+P+((P+P)/(P+P))), 21:(P+P+P+P+P+P+P+P), 22:((P+P+P)/(P+P+P+P)), 23:((P+P+P+P)/(P+P+P)), 24:(P+P+P+P+(P/(P+P))), 25:(P+P+P+((P+P)/(P+P))), 26:((P/(P+P))+(P/(P+P))), 27:((P+P+P+P)/(P+P+P+P)), 28:(P+P+P+P+((P+P)/(P+P))), 29:((P/(P+P))+((P+P)/(P+P))), 30:(((P+P)/(P+P))+((P+P)/(P+P)))]
RERUNNING
.THESE JOBS FAILED: [0:P, 1:(P+P), 2:(P+P+P), 3:(P/(P+P)), 4:(P+P+P+P), 5:(P/(P+P+P)), 6:(P+P+P+P+P), 7:((P+P)/(P+P)), 8:(P+(P/(P+P))), 9:(P/(P+P+P+P)), 10:(P+P+P+P+P+P), 11:((P+P+P)/(P+P)), 12:((P+P)/(P+P+P)), 13:(P+P+(P/(P+P))), 14:(P+((P+P)/(P+P))), 15:(P+P+P+P+P+P+P), 16:((P+P+P)/(P+P+P)), 17:((P+P+P+P)/(P+P)), 18:((P+P)/(P+P+P+P)), 19:(P+P+P+(P/(P+P))), 20:(P+P+((P+P)/(P+P))), 21:(P+P+P+P+P+P+P+P), 22:((P+P+P)/(P+P+P+P)), 23:((P+P+P+P)/(P+P+P)), 24:(P+P+P+P+(P/(P+P))), 25:(P+P+P+((P+P)/(P+P))), 26:((P/(P+P))+(P/(P+P))), 27:((P+P+P+P)/(P+P+P+P)), 28:(P+P+P+P+((P+P)/(P+P))), 29:((P/(P+P))+((P+P)/(P+P))), 30:(((P+P)/(P+P))+((P+P)/(P+P)))]
RERUNNING
.THESE JOBS FAILED: [0:P, 1:(P+P), 2:(P+P+P), 3:(P/(P+P)), 4:(P+P+P+P), 5:(P/(P+P+P)), 6:(P+P+P+P+P), 7:((P+P)/(P+P)), 8:(P+(P/(P+P))), 9:(P/(P+P+P+P)), 10:(P+P+P+P+P+P), 11:((P+P+P)/(P+P)), 12:((P+P)/(P+P+P)), 13:(P+P+(P/(P+P))), 14:(P+((P+P)/(P+P))), 15:(P+P+P+P+P+P+P), 16:((P+P+P)/(P+P+P)), 17:((P+P+P+P)/(P+P)), 18:((P+P)/(P+P+P+P)), 19:(P+P+P+(P/(P+P))), 20:(P+P+((P+P)/(P+P))), 21:(P+P+P+P+P+P+P+P), 22:((P+P+P)/(P+P+P+P)), 23:((P+P+P+P)/(P+P+P)), 24:(P+P+P+P+(P/(P+P))), 25:(P+P+P+((P+P)/(P+P))), 26:((P/(P+P))+(P/(P+P))), 27:((P+P+P+P)/(P+P+P+P)), 28:(P+P+P+P+((P+P)/(P+P))), 29:((P/(P+P))+((P+P)/(P+P))), 30:(((P+P)/(P+P))+((P+P)/(P+P)))]
RERUNNING
.THESE JOBS FAILED: [0:P, 1:(P+P), 2:(P+P+P), 3:(P/(P+P)), 4:(P+P+P+P), 5:(P/(P+P+P)), 6:(P+P+P+P+P), 7:((P+P)/(P+P)), 8:(P+(P/(P+P))), 9:(P/(P+P+P+P)), 10:(P+P+P+P+P+P), 11:((P+P+P)/(P+P)), 12:((P+P)/(P+P+P)), 13:(P+P+(P/(P+P))), 14:(P+((P+P)/(P+P))), 15:(P+P+P+P+P+P+P), 16:((P+P+P)/(P+P+P)), 17:((P+P+P+P)/(P+P)), 18:((P+P)/(P+P+P+P)), 19:(P+P+P+(P/(P+P))), 20:(P+P+((P+P)/(P+P))), 21:(P+P+P+P+P+P+P+P), 22:((P+P+P)/(P+P+P+P)), 23:((P+P+P+P)/(P+P+P)), 24:(P+P+P+P+(P/(P+P))), 25:(P+P+P+((P+P)/(P+P))), 26:((P/(P+P))+(P/(P+P))), 27:((P+P+P+P)/(P+P+P+P)), 28:(P+P+P+P+((P+P)/(P+P))), 29:((P/(P+P))+((P+P)/(P+P))), 30:(((P+P)/(P+P))+((P+P)/(P+P)))]
RERUNNING

... output repeats endlessly

I had to compile the source files myself because the .jar file that is mentioned in the docs (https://github.com/IBM/AI-Descartes/blob/main/symbolic-regression/README.md) does not exist in the repository. I have the demo license of BARON running on my system because I do not want to spend beyond 200$ a month for an academic license, for just trying AI Descartes.

Please let me know how I can run one of the examples on my system.

Recommend Projects