SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1810.12282 · 2018

Assessing Generalization in Deep Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

3 of 6 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; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 0 failures of our instrument; 3 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/1810.12282.json · badge: /api/ran/1810.12282.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
sunblaze-ucb/rl-generalization canonical (classifier label)
classifier basis: census
3 ran · 3 no recorded run
FunctionStatusWhere it lives
ActionDelayWrapper Ran
ran on a synthesized input
sunblaze-ucb/rl-generalization/sunblaze_envs/wrappers.py
code served (permissive licence) · get_code("62d1a513f9887ba4")
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, through the batch API) and returned type. 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 collections
import gym
import numpy as np

# Create a simple test environment
class SimpleEnv(gym.Env):
    def __init__(self):
        self.action_space = gym.spaces.Discrete(4)
        self.observation_space = gym.spaces.Box(low=0, high=1, shape=(4,), dtype=np.float32)
    
    def step(self, action):
        observation = np.random.rand(4).astype(np.float32)
        reward = 1.0
        done = False
        info = {}
        return observation, reward, done, info
    
    def reset(self):
        return np.random.rand(4).astype(np.float32)
    
    def render(self, mode='human'):
        pass

# Set seed for reproducibility
np.random.seed(0)

ARGS = {
    'delay_range_start': 1,
    'delay_range_end': 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
load Ran
ran on a synthesized input
sunblaze-ucb/rl-generalization/sunblaze_envs/registration.py
code served (permissive licence) · get_code("ca03a65a81362f81")
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, through the batch API) and returned module. 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 pkg_resources

# The load function takes a 'name' parameter which is a string
# that gets formatted into an entry point specification.
# It then parses and loads that entry point.
# 
# We need to provide a valid entry point name that can be loaded.
# A simple approach is to use a built-in or standard library module.
# For example, we can use 'os' or 'sys' which are standard modules.

ARGS = {
    'name': 'os'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pkg_resources — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type module — a module output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a module 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
uniform_exclude_inner Ran
ran on a synthesized input
sunblaze-ucb/rl-generalization/sunblaze_envs/classic_control.py
code served (permissive licence) · get_code("67dfd3920059c1d6")
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, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 numpy as np

# Create a numpy uniform random function
np_uniform = np.random.uniform

# Define the outer and inner ranges
a = 0.0
b = 10.0
a_i = 3.0
b_i = 7.0

# Construct the arguments dictionary
ARGS = {
    'np_uniform': np_uniform,
    'a': a,
    'b': b,
    'a_i': a_i,
    'b_i': b_i
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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 Not yet run
unverified (harvested, no recorded run)
sunblaze-ucb/rl-generalization/sunblaze_envs/registration.py
code served (permissive licence) · get_code("8a1139c533d8f08b")
How it ran

no invocation recorded.

register Not yet run
unverified (harvested, no recorded run)
sunblaze-ucb/rl-generalization/sunblaze_envs/registration.py
code served (permissive licence) · get_code("0c4c9f359acd3c78")
How it ran

no invocation recorded.

wrap_environment Not yet run
unverified (harvested, no recorded run)
sunblaze-ucb/rl-generalization/sunblaze_envs/wrappers.py
code served (permissive licence) · get_code("ad77a1249982ee7f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 5 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
register 1 other paper
1 later
2002.06673 (2020-02)
Performative Prediction

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 reinforcement learning (RL) has achieved breakthrough results on many tasks, but agents often fail to generalize beyond the environment they were trained in. As a result, deep RL algorithms that promote generalization are receiving increasing attention. However, works in this area use a wide variety of tasks and experimental setups for evaluation. The literature lacks a controlled assessment of the merits of different generalization schemes. Our aim is to catalyze community-wide progress on generalization in deep RL. To this end, we present a benchmark and experimental protocol, and conduct a systematic empirical study. Our framework contains a diverse set of environments, our methodology covers both in-distribution and out-of-distribution generalization, and our evaluation includes deep RL algorithms that specifically tackle generalization. Our key finding is that `vanilla' deep RL algorithms generalize better than specialized schemes that were proposed specifically to tackle generalization.

For agents

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

get_harvested_code_for_paper("1810.12282")
get_code_for_paper("1810.12282")
have("1810.12282")

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

curl https://syntology.ai/api/ran/1810.12282.json

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

[![Syntology run record](https://syntology.ai/api/ran/1810.12282.svg)](https://syntology.ai/paper/1810.12282)

Connect an agent — have() is free.