SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1712.06567 · 2017

Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 2 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1712.06567.json · badge: /api/ran/1712.06567.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
copy not recorded community
not in the paper's declared repositories
2 ran · 1 no recorded run
optimization-toolbox/DNE4py reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
Bibyutatsu/Self_Driving_Car reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_centered_ranks Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from openai/evolution-strategies-starter
pointer only · get_code("c8c2960f6687b93c")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

compute_ranks Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from openai/evolution-strategies-starter
pointer only · get_code("e2d29b359d6b9487")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

objective Ran
ran, honoured the contract we drafted
optimization-toolbox/DNE4py/DNE4py/version_v2/new_tutorials/tutorials2/pp_run.py
pointer only (licence: GPL-3.0) · get_code("1fc06388f906350c")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, optimization-toolbox/DNE4py, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

uncompress_model Ran
ran; our contract draft was wrong, not the code
Bibyutatsu/Self_Driving_Car/ai.py
code served (permissive licence) · get_code("dac1c1782a9dfe5d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Model. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import random

torch.manual_seed(0)
random.seed(0)

class CompressedModel:
    def __init__(self, start_rng=None, other_rng=None):
        self.start_rng = start_rng if start_rng is not None else random.randint(0, 2 ** 31 - 1)
        self.other_rng = other_rng if other_rng is not None else []

    def evolve(self, sigma, rng_state=None):
        self.other_rng.append((sigma, rng_state if rng_state is not None else random.randint(0, 2 ** 31 - 1)))

class Model(nn.Module):
    def __init__(self, rng_state):
        super().__init__()
        self.hidden = nn.Linear(5, 20)
        self.output = nn.Linear(20, 2)
        self.sigmoid = nn.Sigmoid()
        self.rng_state = rng_state
        torch.manual_seed(rng_state)
        self.evolve_states = []
        self.add_tensors = {}
        for name, tensor in self.named_parameters():
            if tensor.size() not in self.add_tensors:
                self.add_tensors[tensor.size()] = torch.Tensor(tensor.size())
            if 'weight' in name:
                nn.init.kaiming_normal_(tensor)
            else:
                tensor.data.zero_()

    def forward(self, x):
        x = self.hidden(x)
        x = self.sigmoid(x)
        x = self.output(x)
        return x

    def evolve(self, sigma, rng_state):
        torch.manual_seed(rng_state)
        self.evolve_states.append((sigma, rng_state))
        for name, tensor in sorted(self.named_parameters()):
            to_add = self.add_tensors[tensor.size()]
            to_add.normal_(0.0, sigma)
            tensor.data.add_(to_add)

    def compress(self):
        return CompressedModel(self.rng_state, self.evolve_states)

# Create a CompressedModel instance with some evolution history
compressed_model = CompressedModel(
    start_rng=42,
    other_rng=[(0.1, 100), (0.2, 200)]
)

ARGS = {
    "model": compressed_model
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Model — a Model output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Model output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
make_session Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from openai/evolution-strategies-starter
pointer only · get_code("703807555aa85c6d")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 3 the same month, 3 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
compute_centered_ranks 4 other papers
1 dated earlier, 1 the same month, 2 later
1703.03864 (2017-03)
Evolution Strategies as a Scalable Alternative to Reinforcement Learning
1712.06560 (2017-12)
Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents
and 1 more not named here
compute_ranks 4 other papers
1 dated earlier, 1 the same month, 2 later
1703.03864 (2017-03)
Evolution Strategies as a Scalable Alternative to Reinforcement Learning
1712.06560 (2017-12)
Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents
and 1 more not named here
make_session 2 other papers
1 dated earlier, 1 the same month
1703.03864 (2017-03)
Evolution Strategies as a Scalable Alternative to Reinforcement Learning
1712.06560 (2017-12)
Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents
uncompress_model 1 other paper
1 later
1905.04100 (2019-05)
Deep Reinforcement Learning using Genetic Algorithm for Parameter Optimization

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Deep artificial neural networks (DNNs) are typically trained via gradient-based learning algorithms, namely backpropagation. Evolution strategies (ES) can rival backprop-based algorithms such as Q-learning and policy gradients on challenging deep reinforcement learning (RL) problems. However, ES can be considered a gradient-based algorithm because it performs stochastic gradient descent via an operation similar to a finite-difference approximation of the gradient. That raises the question of whether non-gradient-based evolutionary algorithms can work at DNN scales. Here we demonstrate they can: we evolve the weights of a DNN with a simple, gradient-free, population-based genetic algorithm (GA) and it performs well on hard deep RL problems, including Atari and humanoid locomotion. The Deep GA successfully evolves networks with over four million free parameters, the largest neural networks ever evolved with a traditional evolutionary algorithm. These results (1) expand our sense of the scale at which GAs can operate, (2) suggest intriguingly that in some cases following the gradient is not the best choice for optimizing performance, and (3) make immediately available the multitude of neuroevolution techniques that improve performance. We demonstrate the latter by showing that combining DNNs with novelty search, which encourages exploration on tasks with deceptive or sparse reward functions, can solve a high-dimensional problem on which reward-maximizing algorithms (e.g.\ DQN, A3C, ES, and the GA) fail. Additionally, the Deep GA is faster than ES, A3C, and DQN (it can train Atari in ${\raise.17ex\hbox{$\scriptstyle\sim$}}$4 hours on one desktop or ${\raise.17ex\hbox{$\scriptstyle\sim$}}$1 hour distributed on 720 cores), and enables a state-of-the-art, up to 10,000-fold compact encoding technique.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1712.06567")
get_code_for_paper("1712.06567")
have("1712.06567")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/1712.06567.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/1712.06567.svg)](https://syntology.ai/paper/1712.06567)

Connect an agent — have() is free.