SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1908.01289 · 2019

Dueling Posterior Sampling for Preference-Based Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 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; 2 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/1908.01289.json · badge: /api/ran/1908.01289.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
ernovoseller/DuelingPosteriorSampling canonical (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
FunctionStatusWhere it lives
feedback_GPR Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ernovoseller/DuelingPosteriorSampling/Learning_algorithms/DPS_GPR.py
code served (permissive licence) · get_code("67697c55b4941539")
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 dict. 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 dimensions: d=4 state/action pairs, n=3 observations
d = 4
n = 3

# Create GP prior
prior_mean = np.random.randn(d)
prior_cov = np.eye(d) + 0.1 * np.random.randn(d, d)
prior_cov = (prior_cov + prior_cov.T) / 2  # Make symmetric
prior_cov += 0.5 * np.eye(d)  # Ensure positive definite
GP_prior = {
    'mean': prior_mean,
    'cov': prior_cov
}

# Create observation matrix (n-by-d)
observation_matrix = np.random.randn(n, d)

# Create preference labels (length-n vector)
preference_labels = np.array([1.0, -0.5, 0.3])

# Observation noise
obs_noise = 1e-05

ARGS = {
    'GP_prior': GP_prior,
    'observation_matrix': observation_matrix,
    'preference_labels': preference_labels,
    'obs_noise': obs_noise
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
preference_GP_objective Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ernovoseller/DuelingPosteriorSampling/Learning_algorithms/DPS_GP_preference.py
code served (permissive licence) · get_code("4035581cce9a4e17")
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 float64, 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 numpy as np

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

# Construct arguments for preference_GP_objective
d = 4  # number of state/action pairs
n = 2  # number of observations/preferences

# r: length-d vector (the point at which to evaluate the objective)
r = np.random.randn(d)

# observation_matrix: n x d matrix where each row is an observation
observation_matrix = np.random.randn(n, d)

# GP_prior_cov_inv: d x d inverse covariance matrix
# Make it symmetric positive definite
A = np.random.randn(d, d)
GP_prior_cov_inv = A @ A.T + np.eye(d)

# preference_noise: scalar hyperparameter
preference_noise = 0.5

ARGS = {
    "r": r,
    "observation_matrix": observation_matrix,
    "GP_prior_cov_inv": GP_prior_cov_inv,
    "preference_noise": preference_noise,
}
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 float64, 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
feedback_GP_preference Not yet run
unverified (harvested, no recorded run)
ernovoseller/DuelingPosteriorSampling/Learning_algorithms/DPS_GP_preference.py
code served (permissive licence) · get_code("2d292fda4563d367")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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.

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

In preference-based reinforcement learning (RL), an agent interacts with the environment while receiving preferences instead of absolute feedback. While there is increasing research activity in preference-based RL, the design of formal frameworks that admit tractable theoretical analysis remains an open challenge. Building upon ideas from preference-based bandit learning and posterior sampling in RL, we present DUELING POSTERIOR SAMPLING (DPS), which employs preference-based posterior sampling to learn both the system dynamics and the underlying utility function that governs the preference feedback. As preference feedback is provided on trajectories rather than individual state-action pairs, we develop a Bayesian approach for the credit assignment problem, translating preferences to a posterior distribution over state-action reward models. We prove an asymptotic Bayesian no-regret rate for DPS with a Bayesian linear regression credit assignment model. This is the first regret guarantee for preference-based RL to our knowledge. We also discuss possible avenues for extending the proof methodology to other credit assignment models. Finally, we evaluate the approach empirically, showing competitive performance against existing baselines.

For agents

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

get_harvested_code_for_paper("1908.01289")
get_code_for_paper("1908.01289")
have("1908.01289")

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

curl https://syntology.ai/api/ran/1908.01289.json

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

[![Syntology run record](https://syntology.ai/api/ran/1908.01289.svg)](https://syntology.ai/paper/1908.01289)

Connect an agent — have() is free.