SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.09236 · ICML · 2026

Reward Learning through Ranking Mean Squared Error

Matthew Taylor, Chaitanya Kharyal, Calarina Muslimani

arXiv · PDF · Open in the Atlas

Code that ran

2 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, 2 violated it and 0 have no contract verdict; 0 failures of our instrument; 4 no recorded run), measured 2026-11-22; 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/2601.09236.json · badge: /api/ran/2601.09236.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
IRLL/R4 canonical (classifier label)
classifier basis: census
2 ran · 4 no recorded run
FunctionStatusWhere it lives
kendalltau_distance Ran
ran, violated the contract we drafted
IRLL/R4/src/fitnessFunctions/kendalltau.py
code served (permissive licence) · get_code("151580d4be848a32")
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; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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
from scipy.stats import kendalltau

# Create two distinct rankings as inputs
# Rankings should be permutations or orderings of items
ranking1 = np.array([0, 1, 2, 3, 4])
ranking2 = np.array([0, 2, 1, 3, 4])

ARGS = {
    "ranking1": ranking1,
    "ranking2": ranking2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
spearmanr_correlation Ran
ran, violated the contract we drafted
IRLL/R4/src/fitnessFunctions/spearmanr.py
code served (permissive licence) · get_code("c33045e32cf798cc")
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; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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
from scipy.stats import spearmanr

torch.manual_seed(0)

# The function takes two rankings and computes Spearman's r correlation
# Rankings should be array-like sequences of values
ranking1 = np.array([1.0, 2.0, 3.0, 4.0, 5.0])
ranking2 = np.array([2.0, 1.0, 4.0, 3.0, 5.0])

ARGS = {
    "ranking1": ranking1,
    "ranking2": ranking2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, torch — 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 — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
create_q_learning_agent Not yet run
unverified (harvested, no recorded run)
IRLL/R4/src/algos/q_learning.py
code served (permissive licence) · get_code("6c83fa8b8a03a8a8")
How it ran

no invocation recorded.

normalize Not yet run
unverified (harvested, no recorded run)
IRLL/R4/src/lossFunctions/rMSE_ot.py
code served (permissive licence) · get_code("84099d3e84482ccf")
How it ran

no invocation recorded.

read_bin_data Not yet run
unverified (harvested, no recorded run)
IRLL/R4/src/utils.py
code served (permissive licence) · get_code("9e43e4bd1f4486b3")
How it ran

no invocation recorded.

read_pickle Not yet run
unverified (harvested, no recorded run)
IRLL/R4/src/utils.py
code served (permissive licence) · get_code("f76d34b2796b3544")
How it ran

no invocation recorded.

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, 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 6 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 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Reward design remains a significant bottleneck in applying reinforcement learning (RL) to realworld problems. A popular alternative is reward learning, where reward functions are inferred from human feedback rather than manually specified. Recent work has proposed learning reward functions from human ratings rather than traditional binary preferences, enabling richer and potentially less cognitively demanding supervision. Building on this paradigm, we introduce a new rating-based RL method, Ranked Return Regression for RL (R4). At its core, R4 uses a novel ranking mean squared error loss that learns from a dataset of trajectory-rating pairs, treating the human-provided discrete ratings (e.g., "bad," "neutral," "good") as ordinal targets. Unlike prior rating-based approaches, R4 offers formal guarantees: its solution set is provably minimal and complete under mild assumptions. Empirically, using both human-provided and simulated ratings, we demonstrate that R4 consistently matches or outperforms existing rating and preference-based RL methods on robotic benchmarks from OpenAI Gym and the DeepMind Control Suite. Code released at https://github.com/IRLL/R4.

For agents

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

get_harvested_code_for_paper("2601.09236")
get_code_for_paper("2601.09236")
have("2601.09236")

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

curl https://syntology.ai/api/ran/2601.09236.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.09236.svg)](https://syntology.ai/paper/2601.09236)

Connect an agent — have() is free.