hendrycks / ss-ood

Self-Supervised Learning for OOD Detection (NeurIPS 2019)

hendrycks/ss-ood Issues

Signs of the two score components for OOD detection
Updated 2 years ago2
training epochs
Updated 2 years ago
Application of the Method on Signal Data
Closed 3 years ago1
Question for Adversarial Attack Implementation
Closed 3 years ago5
Accuracy for Common Corruptions.
Closed 3 years ago2
Can you provide parameters for multiclass_ood/train_auxiliary?
Closed 3 years ago1
Why negative KL for classification loss when doing OOD detection
Closed 3 years ago2
one-class-train.tar can't be downloaded.
Closed 3 years ago1
how can I reproduce your imageNet AUROC?
Closed 3 years ago1
Reproducing CIFAR-10 One-class OOD Detector with OE
Closed 4 years ago4
wrn_prime not found in adversarial robustness
Closed 4 years ago3
Question about logits, pen = model(adv_bx * 2 - 1)
Closed 4 years ago2
Questions about the model
Closed 4 years ago1
Largely deviating AUROC scores in self-implemented MSP baseline (Multi-Class OoD Detection)
Closed 4 years ago1
why bx*2 -1 ?
Closed 4 years ago1
The DataParallel model doesn't accelerate at all
Closed 4 years ago1
Code for CIFAR-10-C and OOD-CIFAR not provided.
Updated 4 years ago
Inconsistent pixel range values
Closed 4 years ago14
No training data provided
Closed 5 years ago2
Question about Adversarial Training + Rotations
Closed 5 years ago2
Question about loss function (train.py code 168 line)
Closed 5 years ago1
Reproducing robustness results for CIFAR-10 via auxilliary rotation task
Closed 5 years ago1
Would you like to share your trained model?
Closed 5 years ago1
what is Wide RevNet in your paper?
Closed 5 years ago1
Could I use this trick in normal model training?
Closed 5 years ago1