SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.03877 · 2020

FoCL: Feature-Oriented Continual Learning for Generative Models

arXiv · PDF · Open in the Atlas

Code that ran

4 of 8 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 3 have no contract verdict; 1 failure of our instrument; 4 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/2003.03877.json · badge: /api/ran/2003.03877.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
nvcuong/variational-continual-learning canonical (classifier label)
classifier basis: census
4 ran · 4 no recorded run
FunctionStatusWhere it lives
bias_variable Ran
ran; our contract draft was wrong, not the code
nvcuong/variational-continual-learning/ddm/alg/cla_models_multihead.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from jerichooconnell/tf_unet, is pointer only (licence: GPL-3.0) · get_code("5d78e1f7a1afb766")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from jerichooconnell/tf_unet, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

k_center Ran
ran on a synthesized input
nvcuong/variational-continual-learning/ddm/alg/coreset.py
code served (permissive licence) · get_code("2105e7ab152c5c25")
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 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

# Create small sample data for k-center clustering
np.random.seed(0)

# x_train: feature matrix (6 samples, 4 features)
x_train = np.random.randn(6, 4).astype(np.float32)

# y_train: labels (6 samples, 1 label each)
y_train = np.random.randint(0, 2, size=(6, 1)).astype(np.float32)

# Initialize empty lists for coreset
x_coreset = []
y_coreset = []

# Coreset size (must be less than x_train.shape[0])
coreset_size = 3

ARGS = {
    "x_coreset": x_coreset,
    "y_coreset": y_coreset,
    "x_train": x_train,
    "y_train": y_train,
    "coreset_size": coreset_size,
}
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 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
small_variable Ran
ran on a synthesized input
nvcuong/variational-continual-learning/ddm/alg/cla_models_multihead.py
code served (permissive licence) · get_code("61077e3c53bacd00")
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 ResourceVariable, ndim 2, dtype family float, shape (2, 4) 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 tensorflow as tf

def small_variable(shape):
    initial = tf.constant(-6.0, shape=shape)
    return tf.Variable(initial)

ARGS = {
    "shape": (2, 4)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResourceVariable, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
update_distance Ran
ran on a synthesized input
nvcuong/variational-continual-learning/ddm/alg/coreset.py
code served (permissive licence) · get_code("0570a33a48749a3d")
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 (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

# Construct arguments for update_distance function
# The function takes:
# - dists: array of distances (1D)
# - x_train: 2D array of training data points
# - current_id: index of current point (integer)

# Create a small training dataset: 3 samples, 4 features each
x_train = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [2.0, 3.0, 4.0, 5.0],
    [3.0, 4.0, 5.0, 6.0]
], dtype=np.float64)

# Initialize dists array with infinity values
# Length should match number of training samples
dists = np.full(x_train.shape[0], np.inf, dtype=np.float64)

# Set current_id to a valid index (0, 1, or 2)
current_id = 0

ARGS = {
    "dists": dists,
    "x_train": x_train,
    "current_id": current_id
}
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 (3,) — shape (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
config Not yet run
unverified (harvested, no recorded run)
nvcuong/variational-continual-learning/dgm/config.py
code served (permissive licence) · get_code("dededff485c32d1d")
How it ran

no invocation recorded.

load_model Not yet run
unverified (harvested, no recorded run)
nvcuong/variational-continual-learning/dgm/load_classifier.py
code served (permissive licence) · get_code("2934a0c8b74e96f9")
How it ran

no invocation recorded.

rand_from_batch Not yet run
unverified (harvested, no recorded run)
nvcuong/variational-continual-learning/ddm/alg/coreset.py
code served (permissive licence) · get_code("413d98b2d04ac103")
How it ran

no invocation recorded.

weight_variable Not yet run
unverified (harvested, no recorded run)
nvcuong/variational-continual-learning/ddm/alg/cla_models_multihead.py
code served (permissive licence) · get_code("00e7f93923732123")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 7 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
bias_variable 5 other papers
5 dated earlier
1505.04597 (2015-05)
U-Net: Convolutional Networks for Biomedical Image Segmentation
1604.07316 (2016-04)
End to End Learning for Self-Driving Cars
1604.07379 (2016-04)
Context Encoders: Feature Learning by Inpainting
and 2 more not named here

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 this paper, we propose a general framework in continual learning for generative models: Feature-oriented Continual Learning (FoCL). Unlike previous works that aim to solve the catastrophic forgetting problem by introducing regularization in the parameter space or image space, FoCL imposes regularization in the feature space. We show in our experiments that FoCL has faster adaptation to distributional changes in sequentially arriving tasks, and achieves the state-of-the-art performance for generative models in task incremental learning. We discuss choices of combined regularization spaces towards different use case scenarios for boosted performance, e.g., tasks that have high variability in the background. Finally, we introduce a forgetfulness measure that fairly evaluates the degree to which a model suffers from forgetting. Interestingly, the analysis of our proposed forgetfulness score also implies that FoCL tends to have a mitigated forgetting for future tasks.

For agents

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

get_harvested_code_for_paper("2003.03877")
get_code_for_paper("2003.03877")
have("2003.03877")

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

curl https://syntology.ai/api/ran/2003.03877.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.03877.svg)](https://syntology.ai/paper/2003.03877)

Connect an agent — have() is free.