Skip to content
/ LCNet Public

Lu, et al. (2023). Reconciling Shared versus Context-Specific Information in a Neural Network Model of Latent Causes.

License

Notifications You must be signed in to change notification settings

qihongl/LCNet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

10 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

LCNet

This is the repo for the following paper:

Lu, Q., Nguyen, T. T., Zhang, Q., Hasson, U., Griffiths, T. L., Zacks, J. M., Gershman, S. J., & Norman, K. A. (2023). Reconciling Shared versus Context-Specific Information in a Neural Network Model of Latent Causes. In arXiv [q-bio.NC]. arXiv. http://arxiv.org/abs/2312.08519

This repo contains the code for Simulation 1 and Simulation 2. The code for simulation 3 is in a separate repo here.

to replicate the results:

For Simulation 1, run the following code under src

python sim-poly.py

For Simulation 2, run the following code under src

python sim-csw.py

directory structure

.
β”œβ”€β”€ LICENSE
β”œβ”€β”€ README.md
└── src
    β”œβ”€β”€ demo                  
    β”‚Β Β  └── ...... 
    β”œβ”€β”€ model                         # model components 
    β”‚Β Β  β”œβ”€β”€ A2C.py
    β”‚Β Β  β”œβ”€β”€ CGRU.py
    β”‚Β Β  β”œβ”€β”€ CRPLSTM.py
    β”‚Β Β  β”œβ”€β”€ CRPNN.py
    β”‚Β Β  β”œβ”€β”€ ContextRep.py
    β”‚Β Β  β”œβ”€β”€ GRUA2C.py
    β”‚Β Β  β”œβ”€β”€ PECRP.py
    β”‚Β Β  β”œβ”€β”€ PEKDECRP.py
    β”‚Β Β  β”œβ”€β”€ PEKNN.py
    β”‚Β Β  β”œβ”€β”€ PEKNNCRP.py
    β”‚Β Β  β”œβ”€β”€ PETracker.py
    β”‚Β Β  β”œβ”€β”€ ShortCut.py
    β”‚Β Β  β”œβ”€β”€ SimpleContext.py
    β”‚Β Β  β”œβ”€β”€ SimplePETracker.py
    β”‚Β Β  β”œβ”€β”€ TabularShortCut.py
    β”‚Β Β  β”œβ”€β”€ TabularShortCutIS.py
    β”‚Β Β  β”œβ”€β”€ __init__.py
    β”‚Β Β  β”œβ”€β”€ anderson91.py
    β”‚Β Β  └── utils.py
    β”œβ”€β”€ task                         # definition for the tasks
    β”‚Β Β  β”œβ”€β”€ _Polynomial.py           # task used in Simulation 1 
    β”‚Β Β  β”œβ”€β”€ _CSW.py                  # task used in Simulation 2 
    β”‚Β Β  β”œβ”€β”€ _ContextualBandit.py
    β”‚Β Β  β”œβ”€β”€ _MixedCSW.py
    β”‚Β Β  β”œβ”€β”€ _SimpleTwoArmBandit.py
    β”‚Β Β  β”œβ”€β”€ _Waves.py
    β”‚Β Β  └── __init__.py
    β”œβ”€β”€ sim-poly.py                  # code for Simulation 1 
    β”œβ”€β”€ sim-csw.py                   # code for Simulation 2 - model training 
    β”œβ”€β”€ sim-mixedcsw.py             
    β”œβ”€β”€ stats.py                     
    β”œβ”€β”€ utils.py                     
    β”œβ”€β”€ vis-group-poly.py            # code for Simulation 2 - result visualization
    └── vis.py

About

Lu, et al. (2023). Reconciling Shared versus Context-Specific Information in a Neural Network Model of Latent Causes.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages