SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1611.02779 · 2016

RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

11 of 21 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 3 failures of our instrument; 10 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/1611.02779.json · badge: /api/ran/1611.02779.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
aliengirlliv/teachable unrelated (classifier label)
classifier basis: classifier_2025
8 ran · 8 no recorded run
Octavio-Pappalardo/RL2-implementation-pytorch reimplementation (classifier label)
classifier basis: classifier_2025
3 ran · 2 no recorded run
FunctionStatusWhere it lives
DiagGaussianActor Ran
ran on a synthesized input · constructed an object; no result was computed
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("11487404dfdc69d8")
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 an instance of its own class, DiagGaussianActor: the run constructed the object and did not call it, so no result was computed. 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
from torch import nn

torch.manual_seed(0)

ARGS = {
    'obs_dim': 4,
    'action_dim': 3,
    'hidden_dim': 8,
    'hidden_depth': 2,
    'log_std_bounds': (-20, 2),
    'discrete': 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 DiagGaussianActor — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a DiagGaussianActor 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
DictList Ran
ran on a synthesized input · constructed an object; no result was computed
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("c18fe21a87357f31")
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 an instance of its own class, DictList: the run constructed the object and did not call it, so no result was computed. 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
from typing import Any, Dict

torch.manual_seed(0)

# Construct a valid argument for DictList.__init__
# The DictList class takes a dictionary where all values are lists of the same length
# Based on the __len__ method, it returns the length of the first value in the dict
# Based on __getitem__, it supports indexing and returns a new DictList with indexed items

# Create a dictionary with lists of the same size
data_dict = {
    "a": [[1, 2], [3, 4], [5, 6]],
    "b": [[7], [8], [9]],
    "c": [{"x": 10}, {"x": 11}, {"x": 12}]
}

ARGS = {
    "arg": data_dict
}
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 DictList — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a DictList 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
FiLM Ran
ran on a synthesized input · constructed an object; no result was computed
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("4b03e68ac41174cf")
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 an instance of its own class, FiLM: the run constructed the object and did not call it, so no result was computed. 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
from torch import nn

torch.manual_seed(0)

# Create a FiLM instance with reasonable dimensions
# in_features: dimension of input feature vector y
# out_features: number of output channels
# in_channels: number of input channels for x
# imm_channels: number of intermediate channels

ARGS = {
    'in_features': 5,
    'out_features': 4,
    'in_channels': 3,
    'imm_channels': 8
}
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 FiLM — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a FiLM 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
ImageEmbedding Ran
ran on a synthesized input · constructed an object; no result was computed
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("5ffb9bf347805c8e")
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 an instance of its own class, ImageEmbedding: the run constructed the object and did not call it, so no result was computed. 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
from torch import nn

torch.manual_seed(0)

class ObsWrapper:
    """Minimal observation wrapper that holds obs tensor."""
    def __init__(self, obs):
        self.obs = obs

# Create a small input tensor
# ImageBOWEmbedding expects inputs with shape (batch, 3, height, width)
# where values are in range [0, max_value)
# max_value = 147 in the ImageEmbedding
batch_size = 2
height = 4
width = 3
num_channels = 3

# Create input with shape (batch, height, width, num_channels) for transpose operations
input_tensor = torch.randint(0, 147, (batch_size, height, width, num_channels)).float()

# Create observation object
obs = ObsWrapper(input_tensor)

ARGS = {}
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 ImageEmbedding — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ImageEmbedding 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
InstrEmbedding Ran
ran on a synthesized input · constructed an object; no result was computed
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("6552221c0b64c499")
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 an instance of its own class, InstrEmbedding: the run constructed the object and did not call it, so no result was computed. 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 MockVocab:
    def __init__(self, size=32):
        self.size = size
    
    def __len__(self):
        return self.size

class MockEnv:
    def __init__(self):
        self._vocab = MockVocab(size=32)
    
    def vocab(self):
        return self._vocab

class MockObs:
    def __init__(self, batch_size=2, seq_len=5, image_channels=128, image_height=4, image_width=3, instr_dim=128):
        self.instr = torch.randint(0, 32, (batch_size, seq_len), dtype=torch.long)
        self.obs = torch.randn(batch_size, image_channels, image_height, image_width)

class MockArgs:
    pass

args = MockArgs()
env = MockEnv()

ARGS = {
    'args': args,
    'env': env,
    'num_modules': 2,
    'image_dim': 128,
    'instr_dim': 128,
}
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 InstrEmbedding — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a InstrEmbedding 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
Logger Ran
ran on a synthesized input · constructed an object; no result was computed
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("35fc183b2b26ff4a")
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 an instance of its own class, Logger: the run constructed the object and did not call it, so no result was computed. 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 tempfile
import os
from collections import defaultdict

class MockKVWriter(object):
    def writekvs(self, kvs):
        pass
    
    def close(self):
        pass

class MockSeqWriter(object):
    def writeseq(self, seq):
        pass
    
    def close(self):
        pass

temp_dir = tempfile.mkdtemp()

ARGS = {
    'dir': temp_dir,
    'output_formats': [MockKVWriter(), MockSeqWriter()],
    'snapshot_mode': 'last',
    'snapshot_gap': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
joblib — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Logger — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Logger 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
Meta_agent_actor Ran
ran on a synthesized input
Octavio-Pappalardo/RL2-implementation-pytorch/agent_and_tbpttPPO.py
code served (permissive licence) · get_code("792b0690a04ce3bc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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
import torch.nn as nn

def layer_init(layer, std=np.sqrt(2), bias_const=0.0):
    torch.nn.init.orthogonal_(layer.weight, std)
    torch.nn.init.constant_(layer.bias, bias_const)
    return layer

class Meta_agent_actor(nn.Module):

    def __init__(self, hidden_size, actions_size, initial_std=1):
        super(Meta_agent_actor, self).__init__()
        self.initial_std = initial_std
        self.actor_mean = nn.Sequential(layer_init(nn.Linear(hidden_size, 128)), nn.Tanh(), layer_init(nn.Linear(128, 128)), nn.Tanh(), layer_init(nn.Linear(128, actions_size), std=0.01))
        self.log_std = nn.Sequential(layer_init(nn.Linear(hidden_size, 128)), nn.Tanh(), layer_init(nn.Linear(128, actions_size), std=0.01))

    def forward(self, hidden_state):
        if isinstance(hidden_state, tuple):
            hidden_state = hidden_state[0]
        action_mean = self.actor_mean(hidden_state)
        action_logstd = self.log_std(hidden_state)
        action_std = torch.exp(action_logstd) * self.initial_std
        return (action_mean, action_std)


INIT_ARGS = {
    'hidden_size': 4,
    'actions_size': 3,
    'initial_std': 1.0
}

ARGS = {
    'hidden_state': torch.randn(2, 4)
}

if __name__ == '__main__':
    model = Meta_agent_actor(**INIT_ARGS)
    output = model(**ARGS)
    print("Output shapes:")
    print(f"  action_mean: {output[0].shape}")
    print(f"  action_std: {output[1].shape}")
    print("Success!")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Meta_agent_critic Ran
ran on a synthesized input
Octavio-Pappalardo/RL2-implementation-pytorch/agent_and_tbpttPPO.py
code served (permissive licence) · get_code("0379751da7748c2a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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
import torch.nn as nn

def layer_init(layer, std=np.sqrt(2), bias_const=0.0):
    torch.nn.init.orthogonal_(layer.weight, std)
    torch.nn.init.constant_(layer.bias, bias_const)
    return layer

class Meta_agent_critic(nn.Module):

    def __init__(self, hidden_size):
        super(Meta_agent_critic, self).__init__()
        self.critic = nn.Sequential(layer_init(nn.Linear(hidden_size, 512)), nn.Tanh(), layer_init(nn.Linear(512, 1), std=1.0))

    def forward(self, hidden_state):
        if isinstance(hidden_state, tuple):
            hidden_state = hidden_state[0]
        value = self.critic(hidden_state)
        return value


INIT_ARGS = {
    'hidden_size': 64
}

ARGS = {
    'hidden_state': torch.randn(2, 64)
}

if __name__ == '__main__':
    model = Meta_agent_critic(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output: {output}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
layer_init Ran
ran; our contract draft was wrong, not the code
Octavio-Pappalardo/RL2-implementation-pytorch/agent_and_tbpttPPO.py
code served (permissive licence) · get_code("e01d3118f70d5b05")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Linear. 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 numpy as np

torch.manual_seed(0)

# Create a simple linear layer with weight and bias attributes
layer = nn.Linear(in_features=4, out_features=3)

ARGS = {
    'layer': layer,
    'std': np.sqrt(2),
    'bias_const': 0.0
}
Setup block drafted by
a language model, Claude Sonnet 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 Linear — a Linear output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Linear 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
merge_dictlists Ran
ran; our contract draft was wrong, not the code
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("2ab0cf0a97c81ba7")
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 DictList. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

class DictList(dict):
    """A dictionary of lists of same size. Dictionary items can be
    accessed using `.` notation and list items using `[]` notation.

    Example:
        >>> d = DictList({"a": [[1, 2], [3, 4]], "b": [[5], [6]]})
        >>> d.a
        [[1, 2], [3, 4]]
        >>> d[0]
        DictList({"a": [1, 2], "b": [5]})
    """
    __setattr__ = dict.__setitem__

    def __getattr__(self, index):
        try:
            return dict.__getitem__(self, index)
        except KeyError:
            raise AttributeError(index)

    def __len__(self):
        return len(next(iter(dict.values(self))))

    def __getitem__(self, index):
        return DictList({key: value[index] for key, value in dict.items(self)})

    def __setitem__(self, index, d):
        try:
            for key, value in d.items():
                dict.__getitem__(self, key)[index] = value
        except:
            setattr(self, index, d)


# Create test DictList objects with different types of values
# First DictList with lists
dictlist_1 = DictList({
    "list_field": [[1, 2], [3, 4]],
    "tensor_field": torch.randn(2, 4),
    "ndarray_field": np.array([[5.0, 6.0], [7.0, 8.0]]),
    "nested_field": DictList({"inner": [[10], [20]]})
})

# Second DictList with same structure
dictlist_2 = DictList({
    "list_field": [[11, 12], [13, 14]],
    "tensor_field": torch.randn(2, 4),
    "ndarray_field": np.array([[15.0, 16.0], [17.0, 18.0]]),
    "nested_field": DictList({"inner": [[30], [40]]})
})

# Create list of DictLists to merge
list_of_dictlists = [dictlist_1, dictlist_2]

ARGS = {
    "list_of_dictlists": list_of_dictlists
}
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 DictList — a DictList output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DictList 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
mlp Ran
ran; our contract draft was wrong, not the code
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("e26d482acf579f3e")
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 = {
    "input_dim": 4,
    "hidden_dim": 8,
    "output_dim": 3,
    "hidden_depth": 2,
    "output_mod": nn.Sigmoid(),
}
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
Agent Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("0a776339f3429848")
How it ran

no invocation recorded.

ImageBOWEmbedding Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("62cdc662994fc6e1")
How it ran

no invocation recorded.

KVWriter Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("94eba517b065a1f9")
How it ran

no invocation recorded.

Outer_loop_action_agent Not yet run
unverified (harvested, no recorded run)
Octavio-Pappalardo/RL2-implementation-pytorch/agent_and_tbpttPPO.py
code served (permissive licence) · get_code("59c573474692344b")
How it ran

no invocation recorded.

PPOAgent Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("26687176ddf9ce39")
How it ran

no invocation recorded.

SeqWriter Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("75894be2cafb4f9a")
How it ran

no invocation recorded.

collect_evaluation_data Not yet run
unverified (harvested, no recorded run)
Octavio-Pappalardo/RL2-implementation-pytorch/evaluation.py
code served (permissive licence) · get_code("a31c91d734c04428")
How it ran

no invocation recorded.

info Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("6665938be4c3f60c")
How it ran

no invocation recorded.

log Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("d39457795cdad002")
How it ran

no invocation recorded.

weight_init Not yet run
unverified (harvested, no recorded run)
aliengirlliv/teachable/algos/ppo.py
code served (permissive licence) · get_code("6a3b5d4c89f6aec7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 16 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 2 the same month, 15 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
layer_init
brick of this paper's assembly: helper
28 other papers
1 the same month, 26 later, 1 undated · 2 only as an assembly brick
1611.05763 (2016-11)
Learning to reinforcement learn
1703.03400 (2017-03)
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
1803.11485 (2018-03)
QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning
and 25 more not named here
KVWriter
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2101.08152 (2021-01)
Rank the Episodes: A Simple Approach for Exploration in Procedurally-Generated Environments
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
SeqWriter
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2101.08152 (2021-01)
Rank the Episodes: A Simple Approach for Exploration in Procedurally-Generated Environments
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
info
brick of this paper's assembly: helper
2 other papers
2 later · 1 only as an assembly brick
2101.08152 (2021-01)
Rank the Episodes: A Simple Approach for Exploration in Procedurally-Generated Environments
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
DiagGaussianActor
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
DictList
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
FiLM
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
ImageBOWEmbedding
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
ImageEmbedding
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
InstrEmbedding
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
Logger
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
collect_evaluation_data 1 other paper
1 the same month
1611.05763 (2016-11)
Learning to reinforcement learn
log
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
merge_dictlists
brick of this paper's assembly: helper
1 other paper
1 later
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
mlp
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice Distillation
weight_init
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2203.11197 (2022-03)
Teachable Reinforcement Learning via Advice 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

Deep reinforcement learning (deep RL) has been successful in learning sophisticated behaviors automatically; however, the learning process requires a huge number of trials. In contrast, animals can learn new tasks in just a few trials, benefiting from their prior knowledge about the world. This paper seeks to bridge this gap. Rather than designing a "fast" reinforcement learning algorithm, we propose to represent it as a recurrent neural network (RNN) and learn it from data. In our proposed method, RL$^2$, the algorithm is encoded in the weights of the RNN, which are learned slowly through a general-purpose ("slow") RL algorithm. The RNN receives all information a typical RL algorithm would receive, including observations, actions, rewards, and termination flags; and it retains its state across episodes in a given Markov Decision Process (MDP). The activations of the RNN store the state of the "fast" RL algorithm on the current (previously unseen) MDP. We evaluate RL$^2$ experimentally on both small-scale and large-scale problems. On the small-scale side, we train it to solve randomly generated multi-arm bandit problems and finite MDPs. After RL$^2$ is trained, its performance on new MDPs is close to human-designed algorithms with optimality guarantees. On the large-scale side, we test RL$^2$ on a vision-based navigation task and show that it scales up to high-dimensional problems.

For agents

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

get_harvested_code_for_paper("1611.02779")
get_code_for_paper("1611.02779")
have("1611.02779")

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

curl https://syntology.ai/api/ran/1611.02779.json

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

[![Syntology run record](https://syntology.ai/api/ran/1611.02779.svg)](https://syntology.ai/paper/1611.02779)

Connect an agent — have() is free.