Neural decoding for BCI: the objectives, principles and the future

The brain computer interface field draws inspirations from neuroscience, but is also deeply influenced by the emerging trends in the deep learning field and the broader machine learning community. In my talk I will provide a review of recent progress in the field, with a focus on the potential uses of unsupervised and self-supervised learning approaches, the benefits of doing modern machine learning as opposed to neuroscience inspired algorithms, and about modern deep learning approaches. I will reflect on the many rigor violations that are holding back progress in the field. And I will reflect on progress in the field towards highly commoditized approaches.