thanks alot ... i want real matlab code of any DBN application but not handwriting i need it if you could help me please thanks in advance

where's the slide, why the schedule is being displayed. haha

English please, (auto translate subtitling is not working on this video)

Great enlighting talk! I agree with all facts and many myths exposed. Facts are all about benefits of DL and myths are mainly about limitations / downsides of DL. That said, Mr Serrà admits himself the reality of myth # 2 «Big Models Require Big Data» based on simple maths facts: the relation between the number of parameters of a model and the number of training samples. But he cheats a bit when he suggests to regularize complex models (using dropouts) which in fact are becoming simpler models. Since the models used are no more «big models» that means the inherent tasks were not complex. Then he suggests data augmentation / amplification which in fact is a mean to add more data. So «Big Models Require Big Data» is not a myth but a reality and there are sometimes solutions and workarounds. Same thing for the myth # 3, «Lack of Explanation / Interpretability», currently DL models are black-boxes but there are promising and fast moving researches to find solutions. I like the idea of using a second DL model to make sense from the results from another one. I agree it will be the future but it's not a reality yet.