Out-of-Distribution Detection on MNIST vs Fashion-MNIST (1000-tasks Permuted)
100OOD Detection AUCBiMU
Evaluation results
| Method | Links | |
|---|---|---|
| BiMUWeight precision=Binary, Network capacity=2000 neurons, Monte Carlo inference samples=52026.05 | 100 | |
| BiMUWeight Type=Binary, Task Bounds=NO, Architecture=100-unit MLP2026.05 | 99 | |
| SGDWeight precision=Real-valued, Network capacity=2000 neurons, Monte Carlo inference samples=Deterministic2026.05 | 98 | |
| SIWeight precision=Real-valued, Network capacity=2000 neurons, Monte Carlo inference samples=Deterministic2026.05 | 96 | |
| MESUWeight Type=Real-valued, Task Bounds=NO, Architecture=100-unit MLP2026.05 | 95 | |
| SGDWeight Type=Real-valued, Task Bounds=NO, Architecture=100-unit MLP2026.05 | 87 | |
| MESUWeight precision=Real-valued, Network capacity=2000 neurons, Monte Carlo inference samples=102026.05 | 86 | |
| BAYESBINNWeight precision=Binary, Network capacity=2000 neurons, Monte Carlo inference samples=52026.05 | 80 | |
| STEWeight Type=Binary, Task Bounds=NO, Architecture=100-unit MLP2026.05 | 69 | |
| EWC O.Weight Type=Real-valued, Task Bounds=YES, Architecture=100-unit MLP2026.05 | 66 | |
| SIWeight Type=Real-valued, Task Bounds=YES, Architecture=100-unit MLP2026.05 | 66 | |
| STEWeight precision=Binary, Network capacity=2000 neurons, Monte Carlo inference samples=Deterministic2026.05 | 61 | |
| BayesBiNNWeight Type=Binary, Task Bounds=YES, Architecture=100-unit MLP2026.05 | 57 | |
| EWC O.Weight precision=Real-valued, Network capacity=2000 neurons, Monte Carlo inference samples=Deterministic2026.05 | 44 | |
| SYN. META.Weight precision=Binary, Network capacity=2000 neurons, Monte Carlo inference samples=Deterministic2026.05 | 0 |