SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.07495 · 2021

Curiosity-Driven Exploration via Latent Bayesian Surprise

arXiv · PDF · Open in the Atlas

Code that ran

9 of 10 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 0 have no contract verdict; 9 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/2104.07495.json · badge: /api/ran/2104.07495.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
mazpie/lbs-exploration canonical (classifier label)
classifier basis: census
9 ran · 1 no recorded run
FunctionStatusWhere it lives
get_conv_net Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/models/common.py
code served (permissive licence) · get_code("35be8e93eeee05c1")
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 Sequential. 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

torch.manual_seed(0)

ARGS = {
    "state_size": 128,
    "use_bn": False,
    "use_ln": False,
    "mnist_convnet": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
get_random_cnn Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/models/common.py
code served (permissive licence) · get_code("ac436b1c63a904a4")
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 Sequential. 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

torch.manual_seed(0)

ARGS = {
    "state_size": 128,
    "use_bn": False,
    "use_ln": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
get_random_mlp Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/models/common.py
code served (permissive licence) · get_code("761a664f42bbe01d")
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 Sequential. 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

torch.manual_seed(0)

ARGS = {
    "obs_size": 4,
    "state_size": 8,
    "use_bn": False,
    "use_ln": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
get_render_func Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/utils.py
code served (permissive licence) · get_code("108dc9c41448855e")
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 method. 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

class MockEnv:
    def render(self, mode='human'):
        return None

class MockVecEnv:
    def __init__(self):
        self.envs = [MockEnv()]

class MockWrapper:
    def __init__(self, env):
        self.env = env

class MockVecWrapper:
    def __init__(self, venv):
        self.venv = venv

# Test case 1: venv with envs attribute
venv_with_envs = MockVecEnv()

# Test case 2: venv with nested venv attribute
venv_nested = MockVecWrapper(MockVecEnv())

# Test case 3: venv with env attribute
venv_with_env = MockWrapper(MockVecEnv())

# For the function execution, we use the simplest case
ARGS = {
    'venv': venv_with_envs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type method — a method output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a method 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
get_state_block Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
mazpie/lbs-exploration/exploration/environments/half_cheetah_sparse.py
code served (permissive licence) · get_code("e426cc24815ec216")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

torch.manual_seed(0)

# The function expects:
# - state: a tensor-like object with at least 2 elements that support .item() method
# - n_blocks: an integer

state = torch.tensor([5.0, 10.0])
n_blocks = 5

ARGS = {
    "state": state,
    "n_blocks": n_blocks
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
init Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from ArnaudFickinger/adversarial-surprise, is pointer only (licence: NONE) · get_code("9de355e93051e4ad")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ArnaudFickinger/adversarial-surprise, and GitHub detected no licence in that repository when it was checked. 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.

rate_buffer Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/environments/magellan_ant.py
code served (permissive licence) · get_code("6f9031055aa08aed")
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 tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

torch.manual_seed(0)

# The function rate_buffer expects state_buffer which is a list of states
# Each state is indexed with [2] and [3], so it needs to be subscriptable
# with at least 4 elements (indices 0, 1, 2, 3)
# state[2] and state[3] are called with .item(), so they should be torch tensors

# Create sample states - each state needs at least 4 elements
# state[2] should be in range (-1, 5) for x values
# state[3] should be in range (-1, 5) for y values

state1 = torch.tensor([0.0, 0.0, 0.5, 0.5])  # x=0.5 (low), y=0.5 (left) -> block 1
state2 = torch.tensor([0.0, 0.0, 2.0, 2.0])  # x=2.0 (mid), y=2.0 (center) -> not in mapping
state3 = torch.tensor([0.0, 0.0, 4.0, 0.5])  # x=4.0 (high), y=0.5 (left) -> block 3
state4 = torch.tensor([0.0, 0.0, 4.0, 4.0])  # x=4.0 (high), y=4.0 (right) -> block 5

state_buffer = [state1, state3, state4]

ARGS = {
    "state_buffer": state_buffer
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
rate_buffer_with_blocks Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/environments/half_cheetah_sparse.py
code served (permissive licence) · get_code("f9bbb8bf4db866b6")
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 tuple. 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 numpy as np

torch.manual_seed(0)

# Create a small state buffer
# The function expects each state to have at least 2 elements (vel and ang_vel)
# and calls .item() on state[0] and state[1]
state_buffer = [
    torch.tensor([1.5, 3.2]),
    torch.tensor([-2.1, 5.7]),
    torch.tensor([0.5, -1.2]),
    torch.tensor([1.5, 3.2]),  # Duplicate to test uniqueness
]

ARGS = {
    "state_buffer": state_buffer,
    "n_blocks": 6,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
unwrap Ran
ran; our contract draft was wrong, not the code
mazpie/lbs-exploration/exploration/envs.py
code served (permissive licence) · get_code("f2e3732f89410bea")
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 SimpleEnv. 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.

class SimpleEnv:
    def __init__(self):
        pass

class WrappedEnvWithUnwrapped:
    def __init__(self, base_env):
        self.unwrapped = base_env

class WrappedEnvWithEnv:
    def __init__(self, base_env):
        self.env = base_env

class WrappedEnvWithLegEnv:
    def __init__(self, base_env):
        self.leg_env = base_env

# Create a simple base environment
base_env = SimpleEnv()

# Create test environments with different wrapping styles
env_with_unwrapped = WrappedEnvWithUnwrapped(base_env)

ARGS = {
    'env': env_with_unwrapped
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SimpleEnv — a SimpleEnv output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleEnv 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
get_state_block Not yet run
unverified (harvested, no recorded run)
mazpie/lbs-exploration/exploration/environments/magellan_ant.py
code served (permissive licence) · get_code("61ad77d7ae8949a8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 3 later) and 7 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
init 34 other papers
12 dated earlier, 1 the same month, 21 later · 5 only as an assembly brick
1705.05363 (2017-05)
Curiosity-driven Exploration by Self-supervised Prediction
1706.02275 (2017-06)
Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
1810.12894 (2018-10)
Exploration by Random Network Distillation
and 31 more not named here
get_render_func 6 other papers
4 dated earlier, 2 later
1905.04640 (2019-05)
Mega-Reward: Achieving Human-Level Play without Extrinsic Rewards
2003.04108 (2020-03)
Stable Policy Optimization via Off-Policy Divergence Regularization
2006.09447 (2020-06)
Agent Modelling under Partial Observability for Deep Reinforcement Learning
and 3 more not named here
unwrap 2 other papers
2 later
2308.04836 (2023-08)
Beyond Surprise: Improving Exploration Through Surprise Novelty
2401.09750 (2024-01)
Exploration and Anti-Exploration with Distributional Random Network Distillation

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

The human intrinsic desire to pursue knowledge, also known as curiosity, is considered essential in the process of skill acquisition. With the aid of artificial curiosity, we could equip current techniques for control, such as Reinforcement Learning, with more natural exploration capabilities. A promising approach in this respect has consisted of using Bayesian surprise on model parameters, i.e. a metric for the difference between prior and posterior beliefs, to favour exploration. In this contribution, we propose to apply Bayesian surprise in a latent space representing the agent's current understanding of the dynamics of the system, drastically reducing the computational costs. We extensively evaluate our method by measuring the agent's performance in terms of environment exploration, for continuous tasks, and looking at the game scores achieved, for video games. Our model is computationally cheap and compares positively with current state-of-the-art methods on several problems. We also investigate the effects caused by stochasticity in the environment, which is often a failure case for curiosity-driven agents. In this regime, the results suggest that our approach is resilient to stochastic transitions.

For agents

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

get_harvested_code_for_paper("2104.07495")
get_code_for_paper("2104.07495")
have("2104.07495")

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

curl https://syntology.ai/api/ran/2104.07495.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.07495.svg)](https://syntology.ai/paper/2104.07495)

Connect an agent — have() is free.