steffen/self-replicating-neural-networks
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Journal TEx Text

Browse Source
This commit is contained in:
Steffen Illium
2021-09-13 16:04:48 +02:00
parent 6c1a964f31
commit 5f6c658068
7 changed files with 109 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
experiments/soup_exp.py
View File

@@ -63,6 +63,23 @@ class SoupExperiment:
net = Net(self.net_input_size, self.net_hidden_size, self.net_out_size, net_name)
self.population.append(net)
def population_self_train(self):
# Self-training each network in the population
for j in range(self.population_size):
net = self.population[j]
for _ in range(self.ST_steps):
net.self_train(1, self.log_step_size, self.net_learning_rate)
def population_attack(self):
# A network attacking another network with a given percentage
if random.randint(1, 100) <= self.attack_chance:
random_net1, random_net2 = random.sample(range(self.population_size), 2)
random_net1 = self.population[random_net1]
random_net2 = self.population[random_net2]
print(f"\n Attack: {random_net1.name} -> {random_net2.name}")
random_net1.attack(random_net2)
def evolve(self):
""" Evolving consists of attacking & self-training. """
@@ -71,19 +88,10 @@ class SoupExperiment:
loop_epochs.set_description("Evolving soup %s" % i)
# A network attacking another network with a given percentage
if random.randint(1, 100) <= self.attack_chance:
random_net1, random_net2 = random.sample(range(self.population_size), 2)
random_net1 = self.population[random_net1]
random_net2 = self.population[random_net2]
print(f"\n Attack: {random_net1.name} -> {random_net2.name}")
random_net1.attack(random_net2)
self.population_attack()
# Self-training each network in the population
for j in range(self.population_size):
net = self.population[j]
for _ in range(self.ST_steps):
net.self_train(1, self.log_step_size, self.net_learning_rate)
self.population_self_train()
# Testing for fixpoints after each batch of ST steps to see relevant data
if i % self.ST_steps == 0: