Metraipolis › archive › dc_22ee5371c684

Backpropagation-free Deep Learning for Time-Series Classification — arXiv:2610.41805

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.41805
License (as archived): CC0-1.0
Archived: 2026-11-02T04:45:27.129Z
SHA-256 of the archived text: 7f50227c870f7fa499716b12a7554aecd8c550832abdf06d451e53dbea27c30c
Stable URL of this page: https://metraipolis.com/documento/dc_22ee5371c684

How to cite

Backpropagation-free Deep Learning for Time-Series Classification — arXiv:2610.41805 — arXiv (en), https://arxiv.org/abs/2610.41805 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_22ee5371c684 (Metraipolis archive dc_22ee5371c684, sha256 7f50227c870f7fa499716b12a7554aecd8c550832abdf06d451e53dbea27c30c)

BibTeX

@misc{dc22ee5371c684,
  title = {Backpropagation-free Deep Learning for Time-Series Classification — arXiv:2610.41805},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41805},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_22ee5371c684; sha256 7f50227c870f7fa499716b12a7554aecd8c550832abdf06d451e53dbea27c30c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41805. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deep neural networks for time series classification (TSC) are trained almost exclusively by error backpropagation, which imposes update locking and the weight-transport problem. Backpropagation-free learning rules such as Direct Feedback Alignment (DFA), Direct Random Target Projection (DRTP), and the diffusion-inspired NoProp relax these constraints but have been evaluated almost exclusively on 2D image benchmarks. We adapt these three learners to univariate TSC with 1D convolutional architectures and present the first systematic study of them across 112 datasets from the UCR time series archive, with one hyperparameter-optimized configuration per method. The optimized learners reach mean F1 scores of 0.76 to 0.79, below ROCKET (0.83) and InceptionTime (0.84). We propose an analysis framework that relates the per-dataset gap to InceptionTime to dataset properties and tests convolutional feature learning by replacing the flatten readout with global average pooling, measured as a difference-in-differences (DD) against a backpropagation control on the identical architecture. Intra-class temporal variability is the dataset property most consistently associated with the gap for all three learners (partial Spearman $\rho$ from $0.28$ to $0.40$). Pooling harms DFA and DRTP more than backpropagation (DD $-0.025$ and $-0.055$), consistent with near-random convolutional features, whereas NoProp tolerates pooling as backpropagation does (DD $+0.013$). Relative to backpropagation on the same architectures, DRTP reduces training floating-point operations (FLOPs) by 33% and trains faster only on series longer than about 700 time steps, DFA yields no computational reduction, and NoProp reduces peak memory by 38% at the cost of 48% longer training.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.41805 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-02T04:42:12.480Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).