Skip to main content
Search

# Autoencoder

7 related articles

Subscribe to updates
  1. arXiv

    OFDM-ISAC drops the either-or: joint power and constellation optimization eases the sensing-communication trade-off in tracking

    In OFDM-ISAC, communications favor waterfilling power with Gaussian-like inputs while sensing favors uniform power and constant-modulus constellations, creating a trade-off in frequency-selective channels. This work treats per-subcarrier power and kurtosis as joint variables and uses an autoencoder to train offline a family of finite constellations conditioned on kurtosis, generated online for each allocation. Simulations show joint optimization clearly beats optimizing power or constellation alone at intermediate SNR, and the pretrained family nearly realizes the predicted allocation gains.

  2. arXiv

    Fluid world models can read physics but not steer it: pre-registered tests split decodable, prominent and steerable

    On two-dimensional decaying Navier–Stokes flow, three pre-registered studies separate three properties often conflated as evidence that a latent has learned physics: decodability, prominence and steerability. A joint-embedding predictive architecture (JEPA) beats an untrained network of the same architecture on three probe targets yet fails the pre-registered criteria for organizing its latent around fluid quantities; autoencoders that render the field make enstrophy prominent in their slowest unsupervised directions, and adding VICReg lowers that prominence in every comparison while raising linear decodability; a routed design yields a label-free loudness dial in all 8 trained models and a spectral-shape dial in only 3 of 5 fresh seeds, below the pre-registered bar of 4.

  3. Solid-state quantum simulators skip hand fitting: an autoencoder reads effective Hamiltonians from transport data

    Effective Hamiltonian parameters of solid-state quantum simulators can now be inferred directly from transport measurements, without labeled Hamiltonian parameters. The approach embeds the S-matrix formalism, which describes scattering, into an autoencoder decoder so the network can only output physically meaningful parameters. In numerical experiments on a triple quantum dot chain, the model accurately infers parameters and generalizes beyond the training domain while staying robust to noisy measurements; its capacity is finite as the parameter space grows broader or more diverse.

  4. arXiv

    No old samples, one autoencoder per class: TRACE cuts retained memory about 98% below TS-ACL

    For behavioral classes that arrive over time, TRACE stores no historical samples and never expands a shared classifier; it feeds a frozen temporal-spectral representation to a small autoencoder per class and labels by reconstruction error. Across five multivariate time-series class-incremental benchmarks it reaches 88.85% five-dataset average final accuracy with 9.17 MiB of retained memory, about 98.26% less memory than TS-ACL, the strongest analytical baseline by average accuracy, and an approximately 33.43% relative accuracy gain on GrabMyo.

  5. bioRxiv

    Skip the cell-type labels: NetMap reads regulatory networks and cell identity together from mixed single-cell data

    Regulatory network inference no longer needs cell types called first. Earlier pipelines clustered cells, annotated types with marker genes, then inferred one network per cluster, so clustering and annotation errors propagated into the network. NetMap infers a gene regulatory network per cell before any clustering, uses GradientShap to attribute network edges to input genes, and reaches clustering scores of 1.0, 0.99 and 0.99 across three PBMC multiome evaluation cases.

  6. bioRxiv

    STING's two activation routes diverge in dynamics: ligand binding clusters fluctuations at the lid, SAVI mutations reorganize around a1b1

    Molecular dynamics trajectories no longer have to be read only through per-residue fluctuations. Low-population, state-dependent coupling signals were previously hard to separate from high-dimensional trajectories. After compressing each residue pair's side-chain heavy-atom mean-distance time series into a 25-dimensional latent space, then clustering and building networks, the ligand-bound state is dominated by lid residue pairs, while the SAVI-associated multi-mutant shifts to an a1b1-enriched, cross-protomer organization involving residue 281.

  7. An autoencoder plus neural ODE surrogate for ASTEC vessel physics compresses 1913 dimensions to 6 and cuts one simulation from about 4.7 hours to about 19 seconds

    Decoupling the ASTEC vessel physics (the CESAR-ICARE coupling), this work builds a surrogate that uses an autoencoder for dimensionality reduction and a neural ODE to advance time in latent space, training one model each on station blackout and loss-of-coolant accident data; it predicts about 80 scalar and field variables simultaneously, rolls out stably for 10k to 50k time steps (about 4 to 40 hours), compresses 1913 degrees of freedom to 6 latent dimensions (about 332x), and produces the full spatio-temporal prediction in under a minute on both CPU and GPU, with LOCA mean times of about 26.5 s on CPU and 18.9 s on GPU versus 16879.8 s for ASTEC's ICARE module alone, roughly a 640x speedup.

You’re all caught up