Machine learning frameworks such as Tensorflow have lowered the entry barrier to working with neural network models, making them accessible to anyone with a computer and an internet connection. A variety of network architectures have been developed to solve very specific problems with the scalability and accuracy, but understanding why we use one over another can be a challenge.
In this talk we will explore real world use cases and apply several key network architectures – Dense, Convolutional (CNN) and Recurrent (RNN) – using Python and Tensorflow. We’ll identify problems these architectures solve and problems they create, helping you to understand which to use and when.
Delegates will hopefully come away from this talk with a broad understanding of simple neural networks, Convolutional Neural Networks and Recurrent Neural networks, the problems they address and the problems they create.
Basic understanding of Machine learning concepts, awareness of Tensorflow, awareness of Python.