Summary of Vi-pann: Harnessing Transfer Learning and Uncertainty-aware Variational Inference For Improved Generalization in Audio Pattern Recognition, by John Fischer et al.
VI-PANN: Harnessing Transfer Learning and Uncertainty-Aware Variational Inference for Improved Generalization in Audio Pattern Recognition
by John Fischer, Marko Orescanin, Eric Eckstrand
First submitted to arxiv on: 10 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Sound (cs.SD); Audio and Speech Processing (eess.AS)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed VI-PANNs are a variant of ResNet-54 pre-trained on AudioSet, leveraging variational inference for Bayesian deep learning models. By transferring knowledge from these VI-PANNs to other acoustic classification tasks, such as ESC-50, UrbanSound8K, and DCASE2013, the study demonstrates the possibility of transferring calibrated uncertainty information along with model knowledge. This allows for better predictive performance in downstream tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary VI-PANNs use variational inference to achieve competitive predictive performance while providing well-calibrated models that offer epistemic uncertainty in predictions. The study pre-trains VI-PANNs on AudioSet, then evaluates the quality of transferred uncertainty information when applying this knowledge to other acoustic classification tasks. |
Keywords
* Artificial intelligence * Classification * Deep learning * Inference * Resnet
