Sameer Farooqui

Client Services Engineer, Databricks

Sameer Farooqui is a Client Services Engineer at Databricks where he focuses on training and curriculum development. Prior to that, he was a freelance big data + NoSQL consultant and trainer. Before freelancing, Sameer was a Systems Architect at Hortonworks, an Emerging Data Platforms Consultant at Accenture R&D and an Enterprise Solutions Specialist at Symantec (VERITAS division).

SESSIONS

Separating Hype from Reality in Deep Learning – continues

Deep Learning is all the rage these days, but where does the reality of what Deep Learning can do end and the media hype begin? In this talk, I will dispel common myths about Deep Learning that are not necessarily true and help you decide whether you should practically use Deep Learning in your software stack. I will begin with a technical overview of common neural network architectures like CNNs, RNNs, GANs and their common use cases like computer vision, language understanding or unsupervised machine learning. Then I'll separate the hype from reality around questions like: - When should you prefer traditional ML systems like scikit learn or Spark.ML instead of Deep Learning? - Do you no longer need to do careful feature extraction and standardization if using Deep Learning? - Do you really need terabytes of data when training neural networks or can you 'steal' pre-trained lower layers from public models by using transfer learning? - How do you decide which activation function (like ReLU, leaky ReLU, ELU, etc) or optimizer (like Momentum, AdaGrad, RMSProp, Adam, etc) to use in your neural network? - Should you randomly initialize the weights in your network or use more advanced strategies like Xavier or He initialization? - How easy is it to overfit/overtrain a neural network and what are the common techniques to ovoid overfitting (like l1/l2 regularization, dropout and early stopping)?

Separating Hype from Reality in Deep Learning

Deep Learning is all the rage these days, but where does the reality of what Deep Learning can do end and the media hype begin? In this talk, I will dispel common myths about Deep Learning that are not necessarily true and help you decide whether you should practically use Deep Learning in your software stack. I will begin with a technical overview of common neural network architectures like CNNs, RNNs, GANs and their common use cases like computer vision, language understanding or unsupervised machine learning. Then I'll separate the hype from reality around questions like: - When should you prefer traditional ML systems like scikit learn or Spark.ML instead of Deep Learning? - Do you no longer need to do careful feature extraction and standardization if using Deep Learning? - Do you really need terabytes of data when training neural networks or can you 'steal' pre-trained lower layers from public models by using transfer learning? - How do you decide which activation function (like ReLU, leaky ReLU, ELU, etc) or optimizer (like Momentum, AdaGrad, RMSProp, Adam, etc) to use in your neural network? - Should you randomly initialize the weights in your network or use more advanced strategies like Xavier or He initialization? - How easy is it to overfit/overtrain a neural network and what are the common techniques to ovoid overfitting (like l1/l2 regularization, dropout and early stopping)?