AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Towards Knowledge-augmented Bayesian Deep Learning For Computer Vision

Wang Ma, Hanjing Wang, Yufei Zhang, Darsha Udayanga, Qiang Ji

CVPR 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Approximate inference in Bayesian neural networks

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Stage 1 knowledge-informed prior on top of adaptive likelihood (ablation)knowledge-augmented BDL · Decoy MNISTNLL: 0.054 -> 0.043Improved
add Stage 1 knowledge-informed prior on top of adaptive likelihood (ablation)knowledge-augmented BDL · Rotated MNISTaccuracy (%): 94.45 -> 94.68Improved
add Stage 1 knowledge-informed prior on top of adaptive likelihood (ablation)knowledge-augmented BDL · FreiHAND 3D hand reconstructionknowledge constraint KC: 0.26 -> 0.02Improved
add Stage 1 knowledge-informed prior on top of adaptive likelihood (ablation)knowledge-augmented BDL · FreiHAND 3D hand reconstructionjoint error EJ (mm): 8.01 -> 7.89Improved
add Stage 1 knowledge-informed prior (ablation)knowledge-augmented BDL · FreiHAND angle violation rateangle violation rate (%): 2.57% -> 0.14%Improved
add adaptive knowledge likelihood during Stage 2 instead of prior-only knowledge (BANANA)Bayesian deep ensemble · Rotated MNISTaccuracy (%): 51.11 -> 94.45Improved
add adaptive knowledge likelihood vs BANANA prior-onlyBayesian deep ensemble · FreiHAND 3D hand reconstructionjoint error EJ (mm): 19.83 -> 8.01Improved
add Stage 1 prior (ablation)knowledge-augmented BDL · Decoy MNIST with 1% to 10% training dataaccuracyImproved
add Stage 1 prior (ablation)knowledge-augmented BDL · Decoy MNIST to KMNIST OODAUROC (%): 96.78 -> 97.10Mixed
change non-informative prior of likelihood-only model from Gaussian to Laplacian or Logistic (ablation)Ours (Likelihood-Only) · Decoy MNISTaccuracy (%): 98.33 -> 98.30 (Laplacian), 98.39 (Logistic)No clear effect
swap ensemble posterior estimator for SGLD, Bayes-Backprop or Laplace approximation (ablation)Ours (Likelihood-Only) · Decoy MNISTaccuracy (%): 98.33 -> 98.45 (SGLD), 97.76 (Bayes-Backprop), 97.82 (LA)No clear effect
more iterations of adaptive likelihood (Alg. 1) (ablation)Ours (Likelihood-Only) · Decoy MNISTaccuracy (%): 97.52 -> 98.28Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy