Dr. Mohsen Hejrati talks about how to tune hyperparameters and optimize neural networks
Neural networks are proven to be extremely effective at solving many complex tasks. However, the performance and efficiency of neural networks training depend heavily on the choice of its hyperparameters, e.g. learning rate, batch size, layer sizes, etc. Tuning these parameters can become even harder when training on cloud and distributed infrastructure.
During this webinar, our CEO and co-founder, Dr. Mohsen Hejrati presents the tools and approaches for understanding and tuning hyperparameters in a systematic way.
To see our library of all past and upcoming webinars, click here.