SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.02832 · ICML · 2020

Maximum Entropy Gain Exploration for Long Horizon Multi-goal Reinforcement Learning

Silviu Pitis, Harris Chan, Jimmy Ba, Stephen Zhao, Bradly Stadie

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 (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has 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/2007.02832.json · badge: /api/ran/2007.02832.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
penn-pal-lab/peg community
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
MEGA Ran
ran on a synthesized input · constructed an object; no result was computed
penn-pal-lab/peg/dreamerv2/goal_picker.py
code served (permissive licence) · get_code("3847351048947c67")
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 an instance of its own class, MEGA: the run constructed the object and did not call it, so no result was computed. 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
import torch
import torch.nn as nn
from collections import defaultdict

torch.manual_seed(0)

# Create a minimal world model
class MinimalWorldModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(4, 4)
    
    def forward(self, x):
        return self.fc(x)

wm = MinimalWorldModel()

# Create a minimal agent
class MinimalAgent:
    def __init__(self, wm):
        self.wm = wm

agent = MinimalAgent(wm)

# Create a minimal replay buffer
class MinimalReplay:
    def __init__(self, ep_length):
        self.ep_length = ep_length
        self.stats = {
            'loaded_episodes': 5,
            'loaded_steps': 100
        }
        # Create sample episodes with state data
        self._complete_eps = {}
        for ep_id in range(5):
            episode_data = {
                'state': np.random.randn(ep_length, 3),
                'action': np.random.randint(0, 4, ep_length),
                'reward': np.random.randn(ep_length)
            }
            self._complete_eps[ep_id] = episode_data
    
    def dataset(self, batch, length):
        # Generator that yields batches
        while True:
            batch_data = []
            for _ in range(batch):
                ep_id = np.random.randint(0, 5)
                t = np.random.randint(0, self.ep_length - length)
                batch_data.append({
                    'state': self._complete_eps[ep_id]['state'][t:t+length],
                    'action': self._complete_eps[ep_id]['action'][t:t+length]
                })
            yield batch_data

ep_length = 10
replay = MinimalReplay(ep_length)

# Action space
act_space = {'action': 4}

# State key
state_key = 'state'

# Observation to goal function
def obs2goal_fn(obs):
    if isinstance(obs, np.ndarray):
        if obs.ndim == 1:
            return obs[:2]  # Project to 2D
        else:
            return obs[..., :2]  # Project last dimension to 2D
    return obs

ARGS = {
    'agent': agent,
    'replay': replay,
    'act_space': act_space,
    'state_key': state_key,
    'ep_length': ep_length,
    'obs2goal_fn': obs2goal_fn,
    'goal_sample_fn': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type MEGA — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a MEGA 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
softmax Ran
ran; our contract draft was wrong, not the code
penn-pal-lab/peg/dreamerv2/goal_picker.py
code served (permissive licence) · get_code("083546dc5ac62494")
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 2, dtype family float, shape (2, 3) 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

# Create a small 2D array for testing
X = np.array([[1.0, 2.0, 3.0],
              [4.0, 5.0, 6.0]], dtype=np.float32)

# Define the ARGS dict with all parameters
ARGS = {
    "X": X,
    "theta": 1.0,
    "axis": None
}
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 2, dtype family float, shape (2, 3) — shape (2, 3) 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, 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 1 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
softmax
brick of this paper's assembly: helper
1 other paper
1 later
2411.02446 (2024-11)
Learning World Models for Unconstrained Goal Navigation

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

What goals should a multi-goal reinforcement learning agent pursue during training in longhorizon tasks? When the desired (test time) goal distribution is too distant to offer a useful learning signal, we argue that the agent should not pursue unobtainable goals. Instead, it should set its own intrinsic goals that maximize the entropy of the historical achieved goal distribution. We propose to optimize this objective by having the agent pursue past achieved goals in sparsely explored areas of the goal space, which focuses exploration on the frontier of the achievable goal set. We show that our strategy achieves an order of magnitude better sample efficiency than the prior state of the art on long-horizon multi-goal tasks including maze navigation and block stacking. 1

For agents

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

get_harvested_code_for_paper("2007.02832")
get_code_for_paper("2007.02832")
have("2007.02832")

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

curl https://syntology.ai/api/ran/2007.02832.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.02832.svg)](https://syntology.ai/paper/2007.02832)

Connect an agent — have() is free.