SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.07789 · NeurIPS · 2021

Safe Reinforcement Learning by Imagining the Near Future

Tengyu Ma, Garrett Thomas, Yuping Luo

arXiv · PDF · Open in the Atlas

Code that ran

1 of 2 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 1 has no contract verdict; 0 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/2202.07789.json · badge: /api/ran/2202.07789.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
gwthomas/Safe-MBPO canonical (classifier label)
classifier basis not recorded on the link
1 ran · 1 no recorded run
FunctionStatusWhere it lives
unbatched_forward Ran
ran on a synthesized input
gwthomas/Safe-MBPO/src/dynamics.py
code served (permissive licence) · get_code("f8629f3f7e563df7")
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 3, dtype family float, shape (2, 2, 5) 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 torch.nn as nn
import torch.nn.functional as F

torch.manual_seed(0)

class BatchedLinear(nn.Module):
    """For efficient MLP ensembles with batched matrix multiplies"""

    def __init__(self, ensemble_size, in_features, out_features, bias=True):
        super().__init__()
        self.ensemble_size = ensemble_size
        self.in_features = in_features
        self.out_features = out_features
        self.weight = nn.Parameter(torch.empty(ensemble_size, out_features, in_features))
        if bias:
            self.bias = nn.Parameter(torch.empty(ensemble_size, out_features))
        else:
            self.register_parameter('bias', None)
        self.reset_parameters()

    def reset_parameters(self):
        has_bias = self.bias is not None
        l = nn.Linear(self.in_features, self.out_features, bias=has_bias)
        for i in range(self.ensemble_size):
            l.reset_parameters()
            self.weight.data[i].copy_(l.weight.data)
            if has_bias:
                self.bias.data[i].copy_(l.bias.data)

    def forward(self, input):
        assert len(input.shape) == 3
        assert input.shape[0] == self.ensemble_size
        return torch.bmm(input, self.weight.transpose(1, 2)) + self.bias.unsqueeze(1)

def unbatched_forward(batched_sequential, input, index):
    for layer in batched_sequential:
        if isinstance(layer, BatchedLinear):
            input = F.linear(input, layer.weight[index], layer.bias[index])
        else:
            input = layer(input)
    return input

# Create a batched sequential model with BatchedLinear and regular layers
ensemble_size = 2
in_features = 4
hidden_features = 3
out_features = 5

batched_linear_1 = BatchedLinear(ensemble_size, in_features, hidden_features, bias=True)
relu = nn.ReLU()
batched_linear_2 = BatchedLinear(ensemble_size, hidden_features, out_features, bias=True)

batched_sequential = nn.Sequential(
    batched_linear_1,
    relu,
    batched_linear_2
)

# Create input tensor: shape (ensemble_size, batch_size, in_features)
input_tensor = torch.randn(ensemble_size, 2, in_features)

# Index to select which ensemble member to use
index = 0

ARGS = {
    'batched_sequential': batched_sequential,
    'input': input_tensor,
    'index': index
}
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 3, dtype family float, shape (2, 2, 5) — shape (2, 2, 5) 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_env Not yet run
unverified (harvested, no recorded run)
gwthomas/Safe-MBPO/src/shared.py
code served (permissive licence) · get_code("25778383e06d1014")
How it ran

no invocation recorded.

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 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
get_env 1 other paper
1 later
2210.07553 (2022-10)
Safe Model-Based Reinforcement Learning with an Uncertainty-Aware Reachability Certificate
unbatched_forward 1 other paper
1 later
2210.07553 (2022-10)
Safe Model-Based Reinforcement Learning with an Uncertainty-Aware Reachability Certificate

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

Safe reinforcement learning is a promising path toward applying reinforcement learning algorithms to real-world problems, where suboptimal behaviors may lead to actual negative consequences. In this work, we focus on the setting where unsafe states can be avoided by planning ahead a short time into the future. In this setting, a model-based agent with a sufficiently accurate model can avoid unsafe states. We devise a model-based algorithm that heavily penalizes unsafe trajectories, and derive guarantees that our algorithm can avoid unsafe states under certain assumptions. Experiments demonstrate that our algorithm can achieve competitive rewards with fewer safety violations in several continuous control tasks.

For agents

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

get_harvested_code_for_paper("2202.07789")
get_code_for_paper("2202.07789")
have("2202.07789")

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

curl https://syntology.ai/api/ran/2202.07789.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.07789.svg)](https://syntology.ai/paper/2202.07789)

Connect an agent — have() is free.