SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1509.01644 · 2015

Reinforcement Learning with Parameterized Actions

arXiv · PDF · Open in the Atlas

Code that ran

6 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; 1 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 1 failure 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/1509.01644.json · badge: /api/ran/1509.01644.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
cycraig/gym-goal community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
cycraig/MP-DQN reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
angle_difference Ran
ran on a synthesized input
cycraig/gym-goal/gym_goal/envs/util.py
code served (permissive licence) · get_code("d3156c39902cb0a0")
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 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 numpy as np

def norm_angle(angle):
    """ Normalize the angle between -pi and pi. """
    while angle > np.pi:
        angle -= 2 * np.pi
    while angle < -np.pi:
        angle += 2 * np.pi
    return angle

def angle_difference(angle1, angle2):
    """ Computes the real difference between angles. """
    return norm_angle(angle1 - angle2)

# Setup: angle_difference takes two angles as arguments
# Angles can be floats, numpy scalars, or numpy arrays
# Let's test with various angle values

angle1_value = 2.5
angle2_value = 1.2

ARGS = {
    "angle1": angle1_value,
    "angle2": angle2_value,
}
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 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
bound Ran
ran on a synthesized input
cycraig/gym-goal/gym_goal/envs/util.py
code served (permissive licence) · get_code("5cade6842144e12c")
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 int, ndim 0, dtype family int, 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 numpy as np

torch.manual_seed(0)

# The bound function takes three parameters: value, lower, and upper
# It compares value against lower and upper bounds and returns the clipped value
# All three parameters are used in simple numeric comparisons

value = 5
lower = 0
upper = 10

ARGS = {
    "value": value,
    "lower": lower,
    "upper": upper
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, 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
bound_vector Ran
ran on a synthesized input
cycraig/gym-goal/gym_goal/envs/util.py
code served (permissive licence) · get_code("636fc5d1bc7b45f9")
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 ndarray, ndim 1, dtype family float, shape (2,) 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

# Setup for bound_vector function
# The function takes a vector (array-like with at least 2 elements) and a maximum value
# It bounds each element of the vector between -maximum and +maximum

vect = np.array([5.5, -3.2])  # A 2D vector with values that will be clipped
maximum = 2.0  # The bounding maximum value

ARGS = {
    "vect": vect,
    "maximum": maximum
}
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 ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) 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
evaluate Ran
ran, honoured the contract we drafted
cycraig/MP-DQN/run_goal_pdqn.py
code served (permissive licence) · get_code("a07358c3feddc64a")
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 ndarray, ndim 1, dtype family float, shape (2,) 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
import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a minimal agent that implements the act method
class MinimalAgent(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def act(self, state):
        # state is a numpy array of shape (4,)
        # Returns: act (int), act_param (np.array), all_action_parameters (any)
        state_tensor = torch.from_numpy(state).unsqueeze(0)
        output = self.linear(state_tensor)
        
        act = 0  # action index
        act_param = np.zeros((2,))  # parameter for action 0
        all_action_parameters = output.detach().numpy()
        
        return act, act_param, all_action_parameters

# Create a minimal environment that implements reset and step
class MinimalEnv:
    def __init__(self):
        self.step_count = 0
        self.max_steps = 5
    
    def reset(self):
        self.step_count = 0
        # Returns: (state, info)
        state = np.array([0.1, 0.2, 0.3, 0.4], dtype=np.float32)
        info = {}
        return state, info
    
    def step(self, action):
        # action is a tuple (act, params) from pad_action
        self.step_count += 1
        
        # Returns: ((state, info), reward, terminal, truncated)
        next_state = np.array([0.2, 0.3, 0.4, 0.5], dtype=np.float32)
        info = {}
        reward = 1.0
        terminal = self.step_count >= self.max_steps
        truncated = False
        
        return (next_state, info), reward, terminal, truncated

agent = MinimalAgent()
env = MinimalEnv()

ARGS = {
    "env": env,
    "agent": agent,
    "episodes": 2
}
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 ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) 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
norm Ran
ran on a synthesized input
cycraig/gym-goal/gym_goal/envs/goal_env.py
code served (permissive licence) · get_code("b2b2c2c929019580")
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 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 math

def norm(vec2d):
    assert len(vec2d) == 2
    return math.sqrt(vec2d[0] * vec2d[0] + vec2d[1] * vec2d[1])

# The function takes a 2-element vector and computes its Euclidean norm
# vec2d needs to support len() and indexing with [0] and [1]
# A simple list or tuple works perfectly

ARGS = {
    "vec2d": [3.0, 4.0]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
pad_action Ran
ran; our contract draft was wrong, not the code
cycraig/MP-DQN/run_goal_pdqn.py
code served (permissive licence) · get_code("01291f67c9d125d4")
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 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

# Construct valid arguments for pad_action function
act = 1  # action index (0, 1, or 2)
act_param = np.array([0.5, 0.3])  # action parameter array

ARGS = {
    "act": act,
    "act_param": act_param
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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

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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 5 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
evaluate 1 other paper
1 later
1810.06394 (2018-10)
Parametrized Deep Q-Networks Learning: Reinforcement Learning with Discrete-Continuous Hybrid Action Space

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

We introduce a model-free algorithm for learning in Markov decision processes with parameterized actions-discrete actions with continuous parameters. At each step the agent must select both which action to use and which parameters to use with that action. We introduce the Q-PAMDP algorithm for learning in these domains, show that it converges to a local optimum, and compare it to direct policy search in the goal-scoring and Platform domains.

For agents

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

get_harvested_code_for_paper("1509.01644")
get_code_for_paper("1509.01644")
have("1509.01644")

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

curl https://syntology.ai/api/ran/1509.01644.json

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

[![Syntology run record](https://syntology.ai/api/ran/1509.01644.svg)](https://syntology.ai/paper/1509.01644)

Connect an agent — have() is free.