Research map / Label-efficient and robust learning / Recognition and long-tail
Sensor-Based Activity Recognition via Learning From Distributions
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| embed segments with RBF kernel mean map (all moment orders) instead of Moment-10 features | SVM · Skoda | micro-F1: 95.24 -> 99.61 | Helped |
| embed segments with RBF kernel mean map instead of Moment-10 features | SVM · WISDM | micro-F1: 57.79 -> 55.87 | Hurt |
| embed segments with RBF kernel mean map instead of best baseline | SVM · PS | micro-F1: 95.95 -> 96.74 | Helped |
| embed segments with RBF kernel mean map instead of Moment-10 features | SVM · HCI | micro-F1: 98.72 -> 100 | Helped |
| use higher orders of moments in hand-crafted features (Moment-1 to Moment-10) | SVM · Skoda | micro-F1: 92.46 -> 95.24 | Helped |
| use RBF kernel for embedding instead of linear kernel (ablation) | SMMAR · Skoda | micro-F1: 91.4300 -> 98.8950 | Helped |
| use sigmoid kernel for embedding (non-characteristic) instead of RBF (ablation) | SMMAR · Skoda | micro-F1: 98.8950 -> 90.4140 | Hurt |
| replace kernel trick with random Fourier features (R-SMMAR, 10<=D<=40) (ablation) | SMMAR · Skoda | micro-F1 and runtime: 99.61 miF at 264 s | Mixed |
| segment-level kernel embedding instead of frame-level SVM | SVM · WISDM | micro-F1: 27.49 -> 55.87 | Helped |
Closest papers by meaning
- Generalizable Sensor-Based Activity Recognition via Categorical Concept Invariant Learning (AAAI 2025 · similarity 0.84)
- HARDVS: Revisiting Human Activity Recognition with Dynamic Vision Sensors (AAAI 2024 · similarity 0.84)
- Scalable Hierarchical Deep Neural Network for Time Series Analysis in Wearable Sensor-based Human Activity Recognition (AAAI 2025 · similarity 0.82)
- MoPFormer: Motion-Primitive Transformer for Wearable-Sensor Activity Recognition (NeurIPS 2025 · similarity 0.82)
- Decomposing and Fusing Intra- and Inter-Sensor Spatio-Temporal Signal for Multi-Sensor Wearable Human Activity Recognition (AAAI 2025 · similarity 0.82)
- CALANet: Cheap All-Layer Aggregation for Human Activity Recognition (NeurIPS 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.