SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.07069 · 2020

Self-Supervised Discovering of Interpretable Features for Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

6 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 0 have no contract verdict; 6 failures of our instrument; 0 no recorded run), measured 2026-11-21; 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/2003.07069.json · badge: /api/ran/2003.07069.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
shiwj16/SSINet canonical (classifier label)
classifier basis: census
6 ran · 0 no recorded run
FunctionStatusWhere it lives
batchnorm Ran
ran; our contract draft was wrong, not the code
shiwj16/SSINet/net_parts/refinenetLW_parts.py
pointer only (licence: MIT); the stored copy get_code serves, from DrSleep/multi-task-refinenet, is pointer only (licence: NOASSERTION) · get_code("f14815b1ec63be18")
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 DrSleep/multi-task-refinenet, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

center_crop Ran
ran; our contract draft was wrong, not the code
shiwj16/SSINet/net_parts/fcdensenet_parts.py
pointer only (licence: MIT); the stored copy get_code serves, from ravidziv/SimplifyingImbalancedTraining, is pointer only (licence: NONE) · get_code("9d5555c9ff725b7f")
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 ravidziv/SimplifyingImbalancedTraining, 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.

conv1x1 Ran
ran; our contract draft was wrong, not the code
shiwj16/SSINet/net_parts/refinenetLW_parts.py
code served (permissive licence) · get_code("91a91c0edd5005d4")
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 Conv2d. 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 = {
    "in_planes": 64,
    "out_planes": 128,
    "stride": 1,
    "bias": 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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
shiwj16/SSINet/net_parts/refinenetLW_parts.py
code served (permissive licence) · get_code("bc359191fdd6e179")
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 Conv2d. 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 = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1,
    "bias": 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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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_layer Ran
ran; our contract draft was wrong, not the code
shiwj16/SSINet/net_parts/deeplabv3_parts.py
code served (permissive licence) · get_code("ca099495da934438")
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 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)

class SimpleBlock(nn.Module):
    expansion = 2
    
    def __init__(self, in_channels, channels, stride=1, dilation=1):
        super().__init__()
        self.conv = nn.Conv2d(in_channels, channels, kernel_size=3, stride=stride, dilation=dilation, padding=1)
        self.bn = nn.BatchNorm2d(channels)
        self.relu = nn.ReLU(inplace=True)
    
    def forward(self, x):
        out = self.conv(x)
        out = self.bn(out)
        out = self.relu(out)
        return out

ARGS = {
    "block": SimpleBlock,
    "in_channels": 3,
    "channels": 4,
    "num_blocks": 2,
    "stride": 1,
    "dilation": 1
}
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 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
normalized_columns_initializer Ran
ran; our contract draft was wrong, not the code
shiwj16/SSINet/src/networks.py
code served (permissive licence) · get_code("96708b203eb62ef3")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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

torch.manual_seed(0)

# The function expects:
# - weights: a tensor with a .size() method that returns dimensions
#   The function uses weights.size() to create output of same shape
#   Then it computes normalization across dimension 1
#   So weights should be 2D: (num_columns, num_features)

weights = torch.randn(2, 4)  # 2D tensor: rows are columns, cols are features

ARGS = {
    "weights": weights,
    "std": 1.0
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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

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, 6 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 3 later) and 0 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
conv3x3 8 other papers
3 dated earlier, 5 later
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
1805.11046 (2018-05)
Scalable Methods for 8-bit Training of Neural Networks
2001.07645 (2020-01)
SAUNet: Shape Attentive U-Net for Interpretable Medical Image Segmentation
and 5 more not named here
center_crop 4 other papers
2 dated earlier, 2 later
1611.09326 (2016-11)
The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation
1902.02476 (2019-02)
A Simple Baseline for Bayesian Uncertainty in Deep Learning
2312.02517 (2023-12)
Simplifying Neural Network Training Under Class Imbalance
and 1 more not named here
batchnorm 2 other papers
2 dated earlier
1809.04766 (2018-09)
Real-Time Joint Semantic Segmentation and Depth Estimation Using Asymmetric Annotations
2001.07645 (2020-01)
SAUNet: Shape Attentive U-Net for Interpretable Medical Image Segmentation
conv1x1 2 other papers
2 dated earlier
1906.04423 (2019-06)
NAS-FCOS: Fast Neural Architecture Search for Object Detection
2001.07645 (2020-01)
SAUNet: Shape Attentive U-Net for Interpretable Medical Image Segmentation
make_layer 2 other papers
2 later
2203.09418 (2022-03)
ZebraPose: Coarse to Fine Surface Encoding for 6DoF Object Pose Estimation
2510.10177 (2025-10)
HccePose(BF): Predicting Front & Back Surfaces to Construct Ultra-Dense 2D-3D Correspondences for Pose Estimation
normalized_columns_initializer 1 other paper
1 dated earlier
1602.01783 (2016-02)
Asynchronous Methods for Deep Reinforcement Learning

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 2026-11-21.

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 recently led to many breakthroughs on a range of complex control tasks. However, the agent's decision-making process is generally not transparent. The lack of interpretability hinders the applicability of RL in safety-critical scenarios. While several methods have attempted to interpret vision-based RL, most come without detailed explanation for the agent's behavior. In this paper, we propose a self-supervised interpretable framework, which can discover interpretable features to enable easy understanding of RL agents even for non-experts. Specifically, a self-supervised interpretable network (SSINet) is employed to produce fine-grained attention masks for highlighting task-relevant information, which constitutes most evidence for the agent's decisions. We verify and evaluate our method on several Atari 2600 games as well as Duckietown, which is a challenging self-driving car simulator environment. The results show that our method renders empirical evidences about how the agent makes decisions and why the agent performs well or badly, especially when transferred to novel scenes. Overall, our method provides valuable insight into the internal decision-making process of vision-based RL. In addition, our method does not use any external labelled data, and thus demonstrates the possibility to learn high-quality mask through a self-supervised manner, which may shed light on new paradigms for label-free vision learning such as self-supervised segmentation and detection.

For agents

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

get_harvested_code_for_paper("2003.07069")
get_code_for_paper("2003.07069")
have("2003.07069")

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

curl https://syntology.ai/api/ran/2003.07069.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.07069.svg)](https://syntology.ai/paper/2003.07069)

Connect an agent — have() is free.