Synthetic
Benchmarks
TaskDatasetSOTA resultTrendResults
Synthetic 20
14.5Proof Length
6
Synthetic Sawtooth process
1.09Average Predictive Density
6
Synthetic Gaussian Processes
2.57Average Predictive Density
6
Synthetic Quadratic Variance (test)
1.777CRPS
6
Synthetic Linear Variance (test)
1.057CRPS
6
Synthetic Quadratic variance
2.49MSE
6
Synthetic Linear variance
1.512MSE
6
Synthetic nD=10000, d=2, dmax=10, 1000 nodes
5.64CI Test Count
6
Synthetic (d=20)
99.91Coverage
6
Synthetic d=20
99.9Coverage
6
Synthetic d=5 (test)
99.9Coverage
6
Synthetic d=5 (test)
99.91Coverage
6
synthetic
15TP Count
6
Synthetic
0.09P^MC
6
Synthetic Step functions deterministic R matrix (test)
0.02MSE
6
Synthetic Parabolas deterministic R matrix (test)
0.002MSE (mean)
6
Synthetic dataset (1/3 labels)
0.014MAE
6
Synthetic (test)
87.6Helpfulness
6
Synthetic a=1 10-fold (test)
0.90895% PI
6
Synthetic a=0 10-fold (test)
98.195% PI Coverage
6
Synthetic
50.4Latency (s)
6
Synthetic (test)
0.476Mean R2
6
Synthetic (test)
48.41PSNR
6
Synthetic Bernoulli shift
4.04Average Error Rate
6
Synthetic Sine shift
5.75Average Error
6