chroma-v ae
Technology:
80098914b3b3bad79b80377751a85430-Supplemental-Conference.pdf
Outline of Appendices Appendix A describes Chroma-V AE's model and training procedure in detail. Appendix B describes the experimental details in Sections 3 and 5. Appendix C contains Appendix D is a statement on the societal impact of our work. Figure 6 depicts Chroma-V AE and its training procedure. Section 3: CelebA (Synthetic Patch) We use standard CNN architecture. Section 5: ColouredMNIST We use standard CNN architecture.
Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks (0.51)
Chroma-V AE: Mitigating Shortcut Learning with Generative Classifiers
Indeed, we can accuse modern machine learning models of emulating the same behaviour.
Technology:
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.95)
- Information Technology > Artificial Intelligence > Natural Language (0.94)
- Information Technology > Artificial Intelligence > Vision (0.93)
- Information Technology > Artificial Intelligence > Representation & Reasoning (0.93)