SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1905.13402 · 2019

Safety Augmented Value Estimation from Demonstrations (SAVED): Safe Deep Model-Based RL for Sparse Cost Robotic Tasks

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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 4 have no contract verdict; 0 failures of our instrument; 1 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/1905.13402.json · badge: /api/ran/1905.13402.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
harryzhangOG/salved community (classifier label)
classifier basis not recorded on the link
4 ran · 1 no recorded run
FunctionStatusWhere it lives
create_read_only Ran
ran on a synthesized input
harryzhangOG/salved/dmbrl/config/default.py
code served (permissive licence) · get_code("738c96c362e9ea1d")
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

def create_read_only(message):
    def read_only(arg):
        raise RuntimeError(message)
    return read_only

ARGS = {
    "message": "This operation is read-only and cannot be modified"
}
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
goal_distance Ran
ran on a synthesized input
harryzhangOG/salved/dmbrl/env/fetch_env.py
code served (permissive licence) · get_code("b692c1c9edaeb844")
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 numpy as np

# Analyze the function:
# - goal_distance takes two parameters: goal_a and goal_b
# - goal_a is indexed with [:3], [5:8], and [3:5]
# - This suggests goal_a should be an array-like with at least 8 elements
# - The function uses np.linalg.norm with axis=-1, suggesting goal_a could be 2D (batch, features)
# - goal_b is not actually used in the function body, but it's a parameter

# Create goal_a: shape (2, 8) - batch size 2, 8 features
goal_a = np.array([
    [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8],
    [0.15, 0.25, 0.35, 0.45, 0.55, 0.65, 0.75, 0.85]
], dtype=np.float32)

# Create goal_b: similar structure (unused but required parameter)
goal_b = np.array([
    [0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],
    [0.25, 0.35, 0.45, 0.55, 0.65, 0.75, 0.85, 0.95]
], dtype=np.float32)

ARGS = {
    "goal_a": goal_a,
    "goal_b": goal_b
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
lqr_gains Ran
ran on a synthesized input
harryzhangOG/salved/dmbrl/env/pointbot.py
code served (permissive licence) · get_code("26f4997bb8785f25")
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 numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Define small matrices for a linear quadratic regulator problem
# State dimension: 4, Control dimension: 3
n_state = 4
n_control = 3
T = 5  # Time horizon

# System dynamics: x_{t+1} = A @ x_t + B @ u_t
A = np.random.randn(n_state, n_state)

# Control input matrix
B = np.random.randn(n_state, n_control)

# State cost matrix (positive definite)
Q = np.eye(n_state)

# Control cost matrix (positive definite)
R = np.eye(n_control)

ARGS = {
    "A": A,
    "B": B,
    "Q": Q,
    "R": R,
    "T": T,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
make_bool Ran
ran on a synthesized input
harryzhangOG/salved/dmbrl/config/default.py
code served (permissive licence) · get_code("22eb5e3f463c48d7")
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 bool, ndim 0, dtype family float, shape () 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)

ARGS = {
    'arg': '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 bool, ndim 0, dtype family float, shape () — shape () 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
process_action Not yet run
unverified (harvested, no recorded run)
harryzhangOG/salved/dmbrl/env/pointbot.py
code served (permissive licence) · get_code("e403f0ed1d843902")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 2 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
create_read_only 1 other paper
1 later
2006.13916 (2020-06)
Off-Dynamics Reinforcement Learning: Training for Transfer with Domain Classifiers
make_bool 1 other paper
1 later
2006.13916 (2020-06)
Off-Dynamics Reinforcement Learning: Training for Transfer with Domain Classifiers
process_action 1 other paper
1 later
2206.14349 (2022-06)
Fleet-DAgger: Interactive Robot Fleet Learning with Scalable Human Supervision

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

Reinforcement learning (RL) for robotics is challenging due to the difficulty in hand-engineering a dense cost function, which can lead to unintended behavior, and dynamical uncertainty, which makes exploration and constraint satisfaction challenging. We address these issues with a new model-based reinforcement learning algorithm, Safety Augmented Value Estimation from Demonstrations (SAVED), which uses supervision that only identifies task completion and a modest set of suboptimal demonstrations to constrain exploration and learn efficiently while handling complex constraints. We then compare SAVED with 3 state-of-the-art model-based and model-free RL algorithms on 6 standard simulation benchmarks involving navigation and manipulation and a physical knot-tying task on the da Vinci surgical robot. Results suggest that SAVED outperforms prior methods in terms of success rate, constraint satisfaction, and sample efficiency, making it feasible to safely learn a control policy directly on a real robot in less than an hour. For tasks on the robot, baselines succeed less than 5% of the time while SAVED has a success rate of over 75% in the first 50 training iterations. Code and supplementary material is available at https://tinyurl.com/saved-rl.

For agents

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

get_harvested_code_for_paper("1905.13402")
get_code_for_paper("1905.13402")
have("1905.13402")

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

curl https://syntology.ai/api/ran/1905.13402.json

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

[![Syntology run record](https://syntology.ai/api/ran/1905.13402.svg)](https://syntology.ai/paper/1905.13402)

Connect an agent — have() is free.