SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.06242 · ICLR · 2023

HYperbolic Self-Paced Learning for Self-Supervised Skeleton-based Action Representations

Luca Franco, Paolo Mandica, Bharti Munjal, Fabio Galasso

arXiv · PDF · Open in the Atlas

Code that ran

4 of 13 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 4 have no contract verdict; 0 failures of our instrument; 9 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/2303.06242.json · badge: /api/ran/2303.06242.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
paolomandica/hysp community
Papers with Code archive
4 ran · 9 no recorded run
FunctionStatusWhere it lives
DropBlockT_1d Ran
ran on a synthesized input
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("180f114b9cbc15ea")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, 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 import stats

INIT_ARGS = {
    'keep_prob': 0.7
}

# Construct input tensor: batch=2, channels=4, time=6, vertices=3
input_tensor = torch.randn(2, 4, 6, 3)

# Construct mask tensor with same spatial dimensions as input after permutation
# mask shape should match (n, c*v, t) = (2, 4*3, 6) = (2, 12, 6)
mask = torch.rand(2, 12, 6)

ARGS = {
    'input': input_tensor,
    'mask': mask
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 Tensor, ndim 4, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
EMA Ran
ran on a synthesized input
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("8ba7eaf2c8a7739a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned NoneType. 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 torch
import torch.nn as nn

class EMA:
    """Exponential moving average."""

    def __init__(self, beta):
        """Initialize the EMA.

        Args:
            beta (float): Decay rate.
        """
        super().__init__()
        self.beta = beta

    def update_average(self, old, new):
        if old is None:
            return new
        return old * self.beta + (1 - self.beta) * new

    def update_moving_average(self, ma_model, current_model):
        for current_params, ma_params in zip(current_model.parameters(), ma_model.parameters()):
            old_weight, up_weight = (ma_params.data, current_params.data)
            ma_params.data = self.update_average(old_weight, up_weight)


INIT_ARGS = {
    'beta': 0.99
}

# Create two simple models for testing update_moving_average
ma_model = nn.Sequential(
    nn.Linear(4, 3),
    nn.ReLU(),
    nn.Linear(3, 2)
)

current_model = nn.Sequential(
    nn.Linear(4, 3),
    nn.ReLU(),
    nn.Linear(3, 2)
)

ARGS = {
    'ma_model': ma_model,
    'current_model': current_model
}

METHOD = 'update_moving_average'

# Construct and call
ema = EMA(**INIT_ARGS)
ema.update_moving_average(**ARGS)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type NoneType — a NoneType output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a NoneType 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Graph Ran
ran on a synthesized input
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("2bbb5e685ad2f3b3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned ndarray, ndim 3, 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.sparse import csr_matrix

INIT_ARGS = {
    'layout': 'openpose',
    'strategy': 'uniform',
    'max_hop': 1,
    'dilation': 1
}

# The Graph class doesn't have a forward/compute method - it just builds adjacency matrices
# So we'll call __str__ to demonstrate it works (returns self.A)
METHOD = '__str__'

# Create input args that match the expected structure
# Since Graph doesn't take input tensors, ARGS will be empty
ARGS = {}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 ndarray, ndim 3, dtype family float — no output shape is stated: observe_signature.py could not observe it (no_inputs)
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Simam_Drop Ran
ran on a synthesized input
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("dd034bb1c67ea073")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 3, 5) 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

class DropBlockT_1d(nn.Module):

    def __init__(self, keep_prob=0.7):
        super(DropBlockT_1d, self).__init__()
        self.keep_prob = keep_prob

    def forward(self, input, mask):
        n, c, t, v = input.size()
        input1 = input.permute(0, 1, 3, 2).contiguous().view(n, c * v, t)
        mask[mask >= self.keep_prob] = 0.0
        mask[mask < self.keep_prob] = 1.0
        return (input1 * mask * mask.numel() / mask.sum()).view(n, c, v, t).permute(0, 1, 3, 2)

class DropBlock_Ske(nn.Module):

    def __init__(self, num_point=25, keep_prob=0.7):
        super(DropBlock_Ske, self).__init__()
        self.keep_prob = keep_prob
        self.num_point = num_point

    def forward(self, input, mask):
        n, c, t, v = input.size()
        mask[mask >= self.keep_prob] = 0.0
        mask[mask < self.keep_prob] = 1.0
        mask = mask.view(n, 1, 1, self.num_point)
        return input * mask * mask.numel() / mask.sum()

class Simam_Drop(nn.Module):

    def __init__(self, num_point=25, e_lambda=0.0001, keep_prob=0.7):
        super(Simam_Drop, self).__init__()
        self.activaton = nn.Sigmoid()
        self.e_lambda = e_lambda
        self.dropSke = DropBlock_Ske(num_point=num_point, keep_prob=keep_prob)
        self.dropT_skip = DropBlockT_1d(keep_prob=keep_prob)

    def forward(self, x):
        NM, C, T, V = x.size()
        num = V * T - 1
        x_minus_mu_square = (x - x.mean(dim=[2, 3], keepdim=True)).pow(2)
        y = x_minus_mu_square / (4 * (x_minus_mu_square.sum(dim=[2, 3], keepdim=True) / num + self.e_lambda)) + 0.5
        att_map = self.activaton(y)
        att_map_s = att_map.mean(dim=[1, 2])
        att_map_t = att_map.permute(0, 1, 3, 2).contiguous().view(NM, C * V, T)
        output = self.dropT_skip(self.dropSke(x, att_map_s), att_map_t)
        return output

INIT_ARGS = {
    'num_point': 5,
    'e_lambda': 0.0001,
    'keep_prob': 0.7
}

ARGS = {
    'x': torch.randn(2, 4, 3, 5)
}

if __name__ == '__main__':
    model = Simam_Drop(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Input shape: {ARGS['x'].shape}")
    print(f"Output shape: {output.shape}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 4, 3, 5) — shape (2, 4, 3, 5) 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
DropBlock_Ske Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("bc2bd6c395543b89")
How it ran

no invocation recorded.

HYSP Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("90735a76fd308958")
How it ran

no invocation recorded.

HyperMapper Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("80f6560d69b89e68")
How it ran

no invocation recorded.

MLP Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("3a7a63e8f839cd58")
How it ran

no invocation recorded.

MobiusLinear Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("3b892a99281475cb")
How it ran

no invocation recorded.

PositiveOnlyLoss Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("45cf805fdfb46a6c")
How it ran

no invocation recorded.

_mobius_matvec Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("bc7290a25a803047")
How it ran

no invocation recorded.

mobius_linear Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("59823bc0371d269d")
How it ran

no invocation recorded.

mobius_matvec Not yet run
unverified (harvested, no recorded run)
paolomandica/hysp/src/net/hysp.py
code served (permissive licence) · get_code("46dd9b384ab6543c")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 later) and 10 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
mobius_matvec
brick of this paper's assembly: helper
2 other papers
2 dated earlier
2101.01600 (2021-01)
Learning the Predictability of the Future
2211.09224 (2022-11)
Are we certain it's anomalous?
MobiusLinear
brick of this paper's assembly: model
1 other paper
1 dated earlier
2101.01600 (2021-01)
Learning the Predictability of the Future
_mobius_matvec
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2101.01600 (2021-01)
Learning the Predictability of the Future

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

Self-paced learning has been beneficial for tasks where some initial knowledge is available, such as weakly supervised learning and domain adaptation, to select and order the training sample sequence, from easy to complex. However its applicability remains unexplored in unsupervised learning, whereby the knowledge of the task matures during training. We propose a novel HYperbolic Self-Paced model (HYSP) for learning skeletonbased action representations. HYSP adopts self-supervision: it uses data augmentations to generate two views of the same sample, and it learns by matching one (named online) to the other (the target). We propose to use hyperbolic uncertainty to determine the algorithmic learning pace, under the assumption that less uncertain samples should be more strongly driving the training, with a larger weight and pace. Hyperbolic uncertainty is a by-product of the adopted hyperbolic neural networks, it matures during training and it comes with no extra cost, compared to the established Euclidean SSL framework counterparts. When tested on three established skeleton-based action recognition datasets, HYSP outperforms the state-of-the-art on PKU-MMD I, as well as on 2 out of 3 downstream tasks on NTU-60 and NTU-120. Additionally, HYSP only uses positive pairs and bypasses therefore the complex and computationally-demanding mining procedures required for the negatives in contrastive techniques. Code is available at https://github.com/paolomandica/HYSP.

For agents

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

get_harvested_code_for_paper("2303.06242")
get_code_for_paper("2303.06242")
have("2303.06242")

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

curl https://syntology.ai/api/ran/2303.06242.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.06242.svg)](https://syntology.ai/paper/2303.06242)

Connect an agent — have() is free.