Working on XAI for complex computer vision tasks. We’ve built a toolkit which provides the following:
1. Heatmaps based on all popular gradient based explainable AI techniques (plus our own) for classification, regression and semantic segmentation tasks.
2. Uncertainty modeling for classification, segmentation and regression tasks.
3. Concept Discovery/ Pattern Discovery (and dependence) for patterns learned within a deep neural network. (Loosely based on TCAV)
4. Using network internals for optimal pruning and model compression.
Send us an email at [email protected] if you’re interested in trying out our toolkit. We offer 30 day free trial period.
1. Heatmaps based on all popular gradient based explainable AI techniques (plus our own) for classification, regression and semantic segmentation tasks.
2. Uncertainty modeling for classification, segmentation and regression tasks.
3. Concept Discovery/ Pattern Discovery (and dependence) for patterns learned within a deep neural network. (Loosely based on TCAV)
4. Using network internals for optimal pruning and model compression.
Send us an email at [email protected] if you’re interested in trying out our toolkit. We offer 30 day free trial period.