steffen/self-replicating-neural-networks

Go to file

Cristian Lenta 9bf37486a0 Update README.md

2021-05-03 06:50:33 +00:00

experiments.py

uploaded my code (not yet 100% finished)

2021-05-03 06:43:53 +00:00

functionalities_test.py

uploaded my code (not yet 100% finished)

2021-05-03 06:43:53 +00:00

main.py

uploaded my code (not yet 100% finished)

2021-05-03 06:43:53 +00:00

network.py

uploaded my code (not yet 100% finished)

2021-05-03 06:43:53 +00:00

README.md

Update README.md

2021-05-03 06:50:33 +00:00

visualization.py

uploaded my code (not yet 100% finished)

2021-05-03 06:43:53 +00:00

README.md

cristian_lenta - BA code

Changes made from last meeting:

batch_size renamed to log_step_size
changed second order fixpoints: now comparing first input with second output, without weight change of the network
removed rounding of the weights (line 99, network.py): need full float precision

Important:

now that we have full float precision, the robustness test is never failing for the noise values of 10^-6, 10^-7, 10^-8, 10^-9. May just be coding mistake, I am not sure yet.