Pith. sign in

REVIEW 8 cited by

Deep Deterministic Uncertainty: A Simple Baseline

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2102.11582 v3 pith:O6VXVOU7 submitted 2021-02-23 cs.LG stat.ML

classification cs.LGstat.ML
keywords uncertaintydeepdeterministicsimplebaselinecomplexcomputationallyepistemic
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Reliable uncertainty from deterministic single-forward pass models is sought after because conventional methods of uncertainty quantification are computationally expensive. We take two complex single-forward-pass uncertainty approaches, DUQ and SNGP, and examine whether they mainly rely on a well-regularized feature space. Crucially, without using their more complex methods for estimating uncertainty, a single softmax neural net with such a feature-space, achieved via residual connections and spectral normalization, *outperforms* DUQ and SNGP's epistemic uncertainty predictions using simple Gaussian Discriminant Analysis *post-training* as a separate feature-space density estimator -- without fine-tuning on OoD data, feature ensembling, or input pre-procressing. This conceptually simple *Deep Deterministic Uncertainty (DDU)* baseline can also be used to disentangle aleatoric and epistemic uncertainty and performs as well as Deep Ensembles, the state-of-the art for uncertainty prediction, on several OoD benchmarks (CIFAR-10/100 vs SVHN/Tiny-ImageNet, ImageNet vs ImageNet-O) as well as in active learning settings across different model architectures, yet is *computationally cheaper*.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Constrained Co-Design for Photonic Bayesian Neural Networks

    cs.LG 2026-08 conditional novelty 6.0 of 10

    Photonic BNN inference is a constrained variational problem; most hardware limits are trainable through, but signed mean range and some scale bounds are hard representational limits that require hardware changes.

  2. A Unified Evaluation Framework for Epistemic Predictions

    cs.LG 2025-01 conditional novelty 6.0 of 10

    A new metric E = KL distance to the nearest credal-set vertex plus λ times non-specificity ranks uncertainty-aware classifiers under a user-chosen accuracy-precision trade-off.

  3. Density-Informed Pseudo-Counts for Calibrated Evidential Deep Learning

    stat.ML 2026-02 conditional novelty 5.0 of 10

    DIP-EDL sets Dirichlet pseudo-counts to the product of marginal input density and learned class probabilities, concentrating on the true label distribution while sending OOD inputs to the prior.

  4. Optimizing $pp\to A\to Z^{*}h\to \ell^+\ell^- b\bar b$ Searches at the LHC in the 2HDM Type-I with Inverted Hierarchy

    hep-ph 2025-08 conditional novelty 4.0 of 10

    In 2HDM Type-I with inverted hierarchy, applying m_ll in 20-50 GeV and m_bb below 100 GeV makes A to Z* h to ll bb stand out against Drell-Yan and ttbar backgrounds.

  5. HQFNN: A Compact Quantum-Fuzzy Neural Network for Accurate Image Classification

    quant-ph 2025-06 reject novelty 4.0 of 10

    HQFNN is a quantum-fuzzy neural network that embeds fuzzy membership and defuzzification in a small simulated quantum circuit, and it is reported to outperform several specialized baselines on five small image datasets.

  6. Uncertainty and Energy based Loss Guided Semi-Supervised Semantic Segmentation

    cs.CV 2025-01 conditional novelty 4.0 of 10

    A semi-supervised semantic segmentation method combining aleatoric uncertainty loss and energy-based loss with union-intersection pseudo-labels gives modest gains over the CPCL baseline on PASCAL VOC and Cityscapes, b...

  7. Probabilistic Modeling of Disparity Uncertainty for Robust and Efficient Stereo Matching

    cs.CV 2024-12 reject novelty 4.0 of 10

    Stereo matching can report separate data and model uncertainty by combining ordinal-regression disparity distributions with a kernel-regression model-uncertainty estimator.

  8. Risk-aware Classification via Uncertainty Quantification

    cs.LG 2024-12 conditional novelty 4.0 of 10

    Risk-aware evidential classifiers that learn a per-sample prior from a misclassification cost matrix lower expected cost on MNIST and CIFAR10 versus cost-sensitive baselines.

Pith tools