February 15, 2005
Object Smoothing
These results reflect object smoothing and learning all the CPTS:
| Layers of | Model | Time-Slice Accuracy | Edit Distance | |
| Smoothing | Primary Activity | Any Activity | ||
| 0 | "single-trained" straw man (prior) | 84.8% (std 6.3). | 90.35% (std 5.0) | 14.7 std(6.9) |
| 1 | "single-trained" straw man (prior) | 87.35% (std 3.8). | 92.95% (std 3.6) | 6.7 std(3.5) |
| 2 | "single-trained" straw man (prior) | 86.33% (std 3.9). | 92.8% (std 3.5) | 8 std(2.3) |
| 0(starting CPT) | "full-trained" straw man (state-transitions) | 77.55% (std 3.3) | 82.4% (std 4.7) | 66.9 (std 9.98) |
| 1 | "full-trained" straw man (state-transitions) | 77.35% (std 2.7) | 82.41% (std 3.9) | 66 (std 11.76) |
| 2 | "full-trained" straw man (state-transitions) | 72.7% (std 3.4) | 80.27% (std 3.19) | 67.9 (std 11.8) |
| 0 | pinned straw man (partial labels) | 76.47% (std 6.4) | 81.32% (std 5.6) | 56 (std 9.5) |
| 1 | pinned straw man (partial labels) | 79.62% (std 3.4) | 84.41% (std 3.2) | 46.4 (std 8.68) |
| 2 | pinned straw man (partial labels) | 78.5% (std 3.25) | 83.94% (std 2.9) | 46.3.4 (std 7.6) |
| 0 | labelled straw man | 86.48% (std 4.1) | 91.57% (std 3.4) | 14.1 (std 5.7) |
| 1 | labelled straw man | 88.27% (std 2.7) | 93.36% (std 1.7) | 9.6 (std 3.4) |
| 2 | labelled straw man | 88.28% (std 2.7) | 93.37% (std 1.7) | 9.3 (std 3.2) |
February 14, 2005
More improvements to Strawman
These results reflect one change: I smoothed the CPTs between EM iterations to eliminate zero entries.
| Model | Time-Slice Accuracy | Edit Distance | |
| Primary Activity | Any Activity | ||
| "single-trained" straw man (prior) | 84.4% (std 6.6). | 90.1% (std 5.1) | 16.2 std(7.4) |
| "full-trained" straw man (state-transitions) | 71.43% (std 5.3) | 77.71% (std 5.7) | 80.3 (std 9.5) |
| pinned straw man (partial labels) | 51.89% (std 6.8) | 53.38% (std 5.9) | 120.2 (std 12.47) |
| labelled straw man | 86.48% (std 4.1) | 91.57% (std 3.4) | 14.1 (std 5.7) |