1 Introduction
-
We discuss the usability and stability of our protocols on a variety of pretext and target tasks (Sect. 6.2).
-
We find that the objective function mismatch can reduce performance on various benchmarks. Specifically, we observe a performance decrease by \(\sim\)0.1–5.0% for Cifar10, Cifar100 and PCam, and up to \(\sim\)25–59% in extreme cases for the 3dshapes dataset (Sect. 6).
2 Related work
2.1 Unsupervised representation learning
2.2 Analyzing unsupervised representation learning
3 Hard metrics mismatch
3.1 Hard objective function mismatch
4 Soft metrics mismatch
4.1 Soft objective function mismatch
5 Experimental setup
6 Evaluation
6.1 Mismatch and convergence
CAE (Cifar10) | DCAE (Cifar10) | CCAE (Cifar100) | CCAE (PCam) | RCAE (PCam) | SCLCAE (3dshapes) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
\(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {MM3}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {MM3}\) | |
Rep. size | ||||||||||||||
2x2x4 | 0.00 | 0.00 | 0.05 | 0.00 | 0.28 | 1.54 | 4.98 | 9.28 | 5.38 | 4.15 | \(-\)22.26 | 26.34 | \(\infty\) | \(-\)7.99 |
2x2x32 | 0.07 | 1.99 | 0.00 | 3.20 | 0.65 | 3.64 | 5.17 | 34.30 | 3.34 | 7.92 | \(-\)21.09 | 12.98 | 36.39 | \(-\)57.67 |
2x2x128 | 0.20 | 10.10 | 0.06 | 5.51 | 0.51 | 0.81 | 0.32 | 0.10 | 1.03 | 4.04 | \(-\)23.47 | 8.14 | 22.65 | \(-\) 66.19 |
2x2x256 | 0.75 | 11.14 | 0.69 | 5.17 | 0.17 | 0.00 | 0.43 | 0.87 | 0.44 | 0.00 | \(-\)27.60 | 6.52 | 27.65 | \(-\)65.77 |
2x2x512 | 0.43 | 5.28 | 0.36 | 1.25 | 0.00 | 0.00 | 0.20 | 0.07 | 0.18 | 0.00 | \(-\) 28.03 | 5.96 | 27.78 | \(-\)63.61 |
2x2x1024 | 0.24 | 0.25 | 0.03 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.09 | 0.00 | \(-\)26.56 | 4.76 | 32.70 | \(-\)57.92 |
Target model | ||||||||||||||
FC | 0.75 | 11.14 | 0.69 | 5.17 | 0.00 | 0.00 | 0.00 | 0.00 | 1.03 | 4.04 | \(-\)23.47 | 6.52 | 27.65 | \(-\)65.77 |
2FC | 0.03 | 5.68 | 0.00 | 5.14 | 0.08 | 0.00 | 0.00 | 0.00 | 0.31 | 0.76 | \(-\)28.56 | 1.84 | 103.30 | \(-\)70.49 |
3FC | 0.03 | 3.94 | 0.00 | 3.17 | 0.12 | 0.02 | 0.08 | 0.00 | 0.37 | 0.61 | \(-\) 29.61 | 0.91 | 258.18 | \(-\) 71.26 |
Augmentations | ||||||||||||||
All | 0.75 | 11.14 | 0.69 | 5.17 | 0.17 | 0.00 | 0.43 | 0.87 | 1.03 | 4.04 | \(-\)23.47 | 6.52 | 27.65 | \(-\) 65.77 |
NoJitter | 0.99 | 10.99 | 0.50 | 2.33 | – | – | – | – | 0.58 | 0.10 | \(-\) 28.60 | 0.00 | 0.01 | \(-\)37.92 |
NoJitterNoFlip | 1.00 | 12.51 | 0.55 | 1.73 | \(-\) | – | – | – | 1.51 | 7.33 | \(-\)10.60 | 0.00 | 0.00 | \(-\)35.95 |
NoFlip | \(-\) | \(-\) | \(-\) | \(-\) | 0.20 | 0.00 | 0.41 | 0.04 | - | – | – | – | – | – |
6.2 Stability
CAE | DCAE | CCAE | RCAE | SCLCAE | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
\(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {MM3}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {MM3}\) | |
Floor_hue | 0.01 | 0.95 | 0.00 | 1.28 | 0.02 | 0.00 | 56.68 | \(\infty\) | 44.67 | 28.18 | 268.27 | \(-\)48.38 |
Wall_hue | 0.02 | 32.03 | 0.00 | 24.43 | 0.10 | 0.00 | 25.17 | \(\infty\) | 7.80 | 0.29 | 0.46 | \(-\) 76.40 |
Object_hue | 0.38 | 22.71 | 0.43 | 24.55 | 1.55 | 0.63 | 59.65 | \(\infty\) | 40.1 | 2.87 | 8.69 | \(-\)73.17 |
Scale | 0.41 | 0.00 | 0.27 | 0.00 | 0.10 | 0.00 | 2.60 | 0.13 | 31.78 | 2.43 | 0.00 | \(-\)44.80 |
Shape | 0.07 | 0.00 | 0.08 | 0.00 | 0.03 | 0.00 | 0.20 | 0.06 | \(-\) 2.48 | 1.67 | 2.16 | \(-\)67.54 |
Orientation | 0.00 | 0.00 | 0.00 | 0.00 | 0.23 | 0.00 | 0.48 | 0.00 | 22.26 | 2.50 | 6.68 | \(-\)9.11 |
Average | 0.15 | 9.28 | 0.13 | 8.21 | 0.34 | 0.11 | 24.13 | \(\infty\) | 24.02 | 6.32 | 47.71 | \(-\) 53.23 |
6.3 Dependence on representation size
6.4 Dependence on target model complexity
6.5 Dependence on augmentations
6.6 Dependence on target task type
6.7 Applying our metrics to ResNet models
ResNet | ||||
---|---|---|---|---|
\(\mathrm {ACC}\) | \(\mathrm {cSM3}\) | \(\mathrm {MOFM}\) | \(\mathrm {MM3}\) | |
RCAE (Cifar10) | \(54.64^{+1.80}_{-2.01}\) | \(3.98^{+1.74}_{-3.60}\) | \(4.87^{+4.42}_{-3.11}\) | \(31.82^{+0.75}_{-0.68}\) |
SCLCAE (PCam) | \(96.25^{+0.44}_{-0.23}\) | \(0.37^{+0.44}_{-0.37}\) | \(0.86^{+1.00}_{-0.60}\) | \(-53.26^{+0.52}_{-0.38}\) |