SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.10876 · 2022

Hyperparameters in Contextual RL are Highly Situational

arXiv · PDF · Open in the Atlas

Code that ran

2 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 0 failures of our instrument; 0 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/2212.10876.json · badge: /api/ran/2212.10876.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
automl-private/crl_hpo canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
eval_model Ran
ran, honoured the contract we drafted
automl-private/crl_hpo/cRL_HPO/run_pb2.py
code served (permissive licence) · get_code("258d6c5cdb4f0623")
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) and returned float, 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
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# Define a minimal model that has a predict method
class SimplePolicy(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(4, 2)
    
    def predict(self, state):
        # state is expected to be a numpy array or tensor
        if isinstance(state, np.ndarray):
            state_tensor = torch.from_numpy(state).float()
        else:
            state_tensor = torch.tensor(state).float()
        
        # Ensure state_tensor is at least 1D
        if state_tensor.dim() == 0:
            state_tensor = state_tensor.unsqueeze(0)
        
        # If 1D, add batch dimension
        if state_tensor.dim() == 1:
            state_tensor = state_tensor.unsqueeze(0)
        
        action_logits = self.fc(state_tensor)
        action = action_logits.argmax(dim=-1).item()
        value = action_logits.mean().item()
        return action, value

# Define a minimal environment that has reset and step methods
class SimpleEnv:
    def __init__(self):
        self.state_dim = 4
        self.step_count = 0
    
    def reset(self):
        self.step_count = 0
        return np.random.randn(self.state_dim).astype(np.float32)
    
    def step(self, action):
        self.step_count += 1
        next_state = np.random.randn(self.state_dim).astype(np.float32)
        reward = float(np.random.randn())
        done = self.step_count >= 10  # Episode length of 10
        info = {}
        return next_state, reward, done, info

# Create instances
model = SimplePolicy()
eval_env = SimpleEnv()
config = {}

ARGS = {
    "model": model,
    "eval_env": eval_env,
    "config": config
}
Setup block drafted by
a language model, Claude Haiku 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 float, 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
eval_model Ran
ran, honoured the contract we drafted
automl-private/crl_hpo/cRL_HPO/play_pb2.py
code served (permissive licence) · get_code("c73e6161af97557b")
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) and returned float, 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
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# Create a minimal model that has a predict method
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def predict(self, state):
        # state is expected to be a numpy array or tensor
        if isinstance(state, np.ndarray):
            state_tensor = torch.from_numpy(state).float()
        else:
            state_tensor = torch.tensor(state).float()
        
        # Ensure state_tensor is 1D or 2D
        if state_tensor.dim() == 0:
            state_tensor = state_tensor.unsqueeze(0)
        if state_tensor.dim() == 1:
            state_tensor = state_tensor.unsqueeze(0)
        
        action_logits = self.linear(state_tensor)
        action = action_logits.argmax(dim=-1).item()
        return action, None

# Create a minimal environment that has reset and step methods
class MinimalEnv:
    def __init__(self):
        self.state_dim = 4
        self.step_count = 0
    
    def reset(self):
        self.step_count = 0
        return np.random.randn(self.state_dim).astype(np.float32)
    
    def step(self, action):
        self.step_count += 1
        next_state = np.random.randn(self.state_dim).astype(np.float32)
        reward = 1.0
        done = self.step_count >= 10  # Episode ends after 10 steps
        info = {}
        return next_state, reward, done, info

model = MinimalModel()
eval_env = MinimalEnv()

ARGS = {
    "model": model,
    "eval_env": eval_env
}
Setup block drafted by
a language model, Claude Haiku 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 float, 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

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, 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 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.

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

Although Reinforcement Learning (RL) has shown impressive results in games and simulation, real-world application of RL suffers from its instability under changing environment conditions and hyperparameters. We give a first impression of the extent of this instability by showing that the hyperparameters found by automatic hyperparameter optimization (HPO) methods are not only dependent on the problem at hand, but even on how well the state describes the environment dynamics. Specifically, we show that agents in contextual RL require different hyperparameters if they are shown how environmental factors change. In addition, finding adequate hyperparameter configurations is not equally easy for both settings, further highlighting the need for research into how hyperparameters influence learning and generalization in RL.

For agents

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

get_harvested_code_for_paper("2212.10876")
get_code_for_paper("2212.10876")
have("2212.10876")

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

curl https://syntology.ai/api/ran/2212.10876.json

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

[![Syntology run record](https://syntology.ai/api/ran/2212.10876.svg)](https://syntology.ai/paper/2212.10876)

Connect an agent — have() is free.