SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.13268 · ICML · 2022

DRIBO: Robust Deep Reinforcement Learning via Multi-View Information Bottleneck

Jiameng Fan, Wenchao Li

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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; 6 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/2102.13268.json · badge: /api/ran/2102.13268.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
bu-depend-lab/dribo canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
BU-DEPEND-Lab/DRIBO marked official in Papers with Code
Papers with Code archive
4 ran · 1 no recorded run
FunctionStatusWhere it lives
flatten_states Ran
ran; our contract draft was wrong, not the code
BU-DEPEND-Lab/DRIBO/DRIBO/DRIBO_sac.py
code served (permissive licence) · get_code("c86249c3d9ba943f")
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 RSSMState. 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
from collections import namedtuple

torch.manual_seed(0)

# First, define the helper function and RSSMState class
RESERVED_NAMES = ('get', 'items')

def tuple_itemgetter(i):
    def _tuple_itemgetter(obj):
        return tuple.__getitem__(obj, i)
    return _tuple_itemgetter

def namedarraytuple(typename, field_names, return_namedtuple_cls=False, classname_suffix=False):
    nt_typename = typename
    if classname_suffix:
        nt_typename += '_nt'
        typename += '_nat'
    try:
        import sys
        module = sys._getframe(1).f_globals.get('__name__', '__main__')
    except (AttributeError, ValueError):
        module = None
    NtCls = namedtuple(nt_typename, field_names, module=module)

    def __getitem__(self, loc):
        try:
            return type(self)(*(None if s is None else s[loc] for s in self))
        except IndexError as e:
            for j, s in enumerate(self):
                if s is None:
                    continue
                try:
                    _ = s[loc]
                except IndexError:
                    raise Exception(f"Occured in {self.__class__} at field '{self._fields[j]}'.") from e
    __getitem__.__doc__ = f'Return a new {typename} instance containing the selected index or slice from each field.'

    def __setitem__(self, loc, value):
        if not (isinstance(value, tuple) and getattr(value, '_fields', None) == self._fields):
            value = tuple((None if s is None else value for s in self))
        try:
            for j, (s, v) in enumerate(zip(self, value)):
                if s is not None or v is not None:
                    s[loc] = v
        except (ValueError, IndexError, TypeError) as e:
            raise Exception(f"Occured in {self.__class__} at field '{self._fields[j]}'.") from e

    def __contains__(self, key):
        return key in self._fields

    def get(self, index):
        return tuple.__getitem__(self, index)

    def items(self):
        for k, v in zip(self._fields, self):
            yield (k, v)
    
    for method in (__getitem__, __setitem__, get, items):
        method.__qualname__ = f'{typename}.{method.__name__}'
    
    arg_list = repr(NtCls._fields).replace("'", '')[1:-1]
    class_namespace = {'__doc__': f'{typename}({arg_list})', '__slots__': (), '__getitem__': __getitem__, '__setitem__': __setitem__, '__contains__': __contains__, 'get': get, 'items': items}
    for index, name in enumerate(NtCls._fields):
        if name in RESERVED_NAMES:
            raise ValueError(f'Disallowed field name: {name}.')
        itemgetter_object = tuple_itemgetter(index)
        doc = f'Alias for field number {index}'
        class_namespace[name] = property(itemgetter_object, doc=doc)
    result = type(typename, (NtCls,), class_namespace)
    result.__module__ = NtCls.__module__
    if return_namedtuple_cls:
        return (result, NtCls)
    return result

RSSMState = namedarraytuple('RSSMState', ['mean', 'std', 'stoch', 'deter'])

# Create test data
rssm_state = RSSMState(
    mean=torch.randn(2, 4, 5),
    std=torch.randn(2, 4, 5),
    stoch=torch.randn(2, 4, 5),
    deter=torch.randn(2, 4, 5)
)

batch_shape = 2

ARGS = {
    'rssm_states': rssm_state,
    'batch_shape': batch_shape,
}
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 RSSMState — a RSSMState output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RSSMState output the recorded ndim 3 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
gaussian_logprob Ran
ran; our contract draft was wrong, not the code
bu-depend-lab/dribo/DRIBO/DRIBO_sac.py
code served (permissive licence) · get_code("daa7b3a0355eb7c3")
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 Tensor, ndim 2, dtype family float, shape (2, 1) 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 torch
import numpy as np

torch.manual_seed(0)

# Create a small noise tensor with shape (batch_size=2, feature_dim=4)
noise = torch.randn(2, 4)

# Create a small log_std tensor with shape (batch_size=2, feature_dim=4)
log_std = torch.randn(2, 4)

ARGS = {
    "noise": noise,
    "log_std": log_std,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 1) — shape (2, 1) 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
get_dist Ran
ran; our contract draft was wrong, not the code
BU-DEPEND-Lab/DRIBO/DRIBO/DRIBO_sac.py
code served (permissive licence) · get_code("6e2c1f59f1d53fcc")
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 Independent. 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.distributions as td
from collections import namedtuple

torch.manual_seed(0)

# First, define the tuple_itemgetter function
def tuple_itemgetter(i):
    def _tuple_itemgetter(obj):
        return tuple.__getitem__(obj, i)
    return _tuple_itemgetter

# Define namedarraytuple (simplified version for our use case)
RESERVED_NAMES = ('get', 'items')

def namedarraytuple(typename, field_names, return_namedtuple_cls=False, classname_suffix=False):
    nt_typename = typename
    if classname_suffix:
        nt_typename += '_nt'
        typename += '_nat'
    
    NtCls = namedtuple(nt_typename, field_names)

    def __getitem__(self, loc):
        try:
            return type(self)(*(None if s is None else s[loc] for s in self))
        except IndexError as e:
            for j, s in enumerate(self):
                if s is None:
                    continue
                try:
                    _ = s[loc]
                except IndexError:
                    raise Exception(f"Occured in {self.__class__} at field '{self._fields[j]}'.") from e

    def __setitem__(self, loc, value):
        if not (isinstance(value, tuple) and getattr(value, '_fields', None) == self._fields):
            value = tuple((None if s is None else value for s in self))
        try:
            for j, (s, v) in enumerate(zip(self, value)):
                if s is not None or v is not None:
                    s[loc] = v
        except (ValueError, IndexError, TypeError) as e:
            raise Exception(f"Occured in {self.__class__} at field '{self._fields[j]}'.") from e

    def __contains__(self, key):
        return key in self._fields

    def get(self, index):
        return tuple.__getitem__(self, index)

    def items(self):
        for k, v in zip(self._fields, self):
            yield (k, v)

    for method in (__getitem__, __setitem__, get, items):
        method.__qualname__ = f'{typename}.{method.__name__}'
    
    arg_list = repr(NtCls._fields).replace("'", '')[1:-1]
    class_namespace = {'__doc__': f'{typename}({arg_list})', '__slots__': (), '__getitem__': __getitem__, '__setitem__': __setitem__, '__contains__': __contains__, 'get': get, 'items': items}
    
    for index, name in enumerate(NtCls._fields):
        if name in RESERVED_NAMES:
            raise ValueError(f'Disallowed field name: {name}.')
        itemgetter_object = tuple_itemgetter(index)
        doc = f'Alias for field number {index}'
        class_namespace[name] = property(itemgetter_object, doc=doc)
    
    result = type(typename, (NtCls,), class_namespace)
    result.__module__ = NtCls.__module__
    
    if return_namedtuple_cls:
        return (result, NtCls)
    return result

# Create RSSMState
RSSMState = namedarraytuple('RSSMState', ['mean', 'std', 'stoch', 'deter'])

# Create valid arguments for get_dist
mean = torch.randn(2, 4)
std = torch.abs(torch.randn(2, 4)) + 0.1
stoch = torch.randn(2, 3)
deter = torch.randn(2, 5)

rssm_state = RSSMState(mean=mean, std=std, stoch=stoch, deter=deter)

ARGS = {
    'rssm_state': rssm_state
}
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 Independent — a Independent output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Independent 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
namedarraytuple Ran
ran; our contract draft was wrong, not the code
BU-DEPEND-Lab/DRIBO/DRIBO/DRIBO_sac.py
code served (permissive licence) · get_code("ead01b9035a66905")
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 numpy as np
import torch
import sys
from collections import namedtuple

torch.manual_seed(0)

ARGS = {
    'typename': 'TestTuple',
    'field_names': ['data', 'labels'],
    'return_namedtuple_cls': False,
    'classname_suffix': False,
}
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 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
squash Ran
ran; our contract draft was wrong, not the code
bu-depend-lab/dribo/DRIBO/DRIBO_sac.py
code served (permissive licence) · get_code("e4f442d1770eacd1")
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 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 torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for the function arguments
batch_size = 2
dim = 4

# mu: input tensor that will be passed through tanh
mu = torch.randn(batch_size, dim)

# pi: tensor that will be passed through tanh (can be None)
pi = torch.randn(batch_size, dim)

# log_pi: log probability tensor (can be None)
log_pi = torch.randn(batch_size, 1)

ARGS = {
    "mu": mu,
    "pi": pi,
    "log_pi": log_pi,
}
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 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
tuple_itemgetter Ran
ran; our contract draft was wrong, not the code
BU-DEPEND-Lab/DRIBO/DRIBO/DRIBO_sac.py
code served (permissive licence) · get_code("0fe5032c8a208099")
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 function. 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

torch.manual_seed(0)

def tuple_itemgetter(i):
    def _tuple_itemgetter(obj):
        return tuple.__getitem__(obj, i)
    return _tuple_itemgetter

# Test the function to understand what it needs
# The function takes an index i and returns a function that gets the i-th item from a tuple
test_tuple = (10, 20, 30)
getter = tuple_itemgetter(1)
result = getter(test_tuple)

ARGS = {
    'i': 0
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
DRIBO Not yet run
unverified (harvested, no recorded run)
BU-DEPEND-Lab/DRIBO/DRIBO/DRIBO_sac.py
code served (permissive licence) · get_code("54e2e388a1b71f60")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 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
gaussian_logprob 15 other papers
7 dated earlier, 8 later · 3 only as an assembly brick
1910.01741 (2019-10)
Improving Sample Efficiency in Model-Free Reinforcement Learning from Images
2004.04136 (2020-04)
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
and 12 more not named here
squash 14 other papers
7 dated earlier, 7 later · 3 only as an assembly brick
1910.01741 (2019-10)
Improving Sample Efficiency in Model-Free Reinforcement Learning from Images
2004.04136 (2020-04)
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
and 11 more not named here

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 (DRL) agents are often sensitive to visual changes that were unseen in their training environments. To address this problem, we leverage the sequential nature of RL to learn robust representations that encode only task-relevant information from observations based on the unsupervised multi-view setting. Specifically, we introduce a novel contrastive version of the Multi-View Information Bottleneck (MIB) objective for temporal data. We train RL agents from pixels with this auxiliary objective to learn robust representations that can compress away task-irrelevant information and are predictive of task-relevant dynamics. This approach enables us to train high-performance policies that are robust to visual distractions and can generalize well to unseen environments. We demonstrate that our approach can achieve SOTA performance on a diverse set of visual control tasks in the DeepMind Control Suite when the background is replaced with natural videos. In addition, we show that our approach outperforms well-established baselines for generalization to unseen environments on the Procgen benchmark. Our code is opensourced and available at https://github. com/BU-DEPEND-Lab/DRIBO.

For agents

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

get_harvested_code_for_paper("2102.13268")
get_code_for_paper("2102.13268")
have("2102.13268")

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

curl https://syntology.ai/api/ran/2102.13268.json

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

[![Syntology run record](https://syntology.ai/api/ran/2102.13268.svg)](https://syntology.ai/paper/2102.13268)

Connect an agent — have() is free.