Ομιλία του Καθηγητή Richard Baraniuk από το Rice University, USA, με θέμα: "A Probabilistic Theory of Deep Learning", στο ΕΜΠ

<p>A grand challenge in machine learning is the development of computational algorithms that match or outperform humans in perceptual inference tasks that are complicated by nuisance variation. For instance, visual object recognition involves the unknown object position, orientation, and scale in object recognition while speech recognition involves the unknown voice pronunciation, pitch, and speed. Recently, a new breed of deep learning algorithms have emerged for high-nuisance inference tasks that routinely yield pattern recognition systems with near- or super-human capabilities. However, a coherent framework for understanding, analyzing, and synthesizing deep learning architectures has remained elusive. We make some progress in this direction by developing a new probabilistic framework for deep learning based on the Deep Rendering Model (DRM): a generative probabilistic model that explicitly captures latent nuisance variation. By relaxing the DRM&rsquo;s generative model to a discriminative one, we recover the inference computations in not only deep convolutional neural networks but also random decision forests, providing insights into their successes and shortcomings, a principled route to their improvement, and new avenues for exploration. (2016 NIPS paper: https://goo.gl/kNcXG1)</p>