Deep learning has come a long way since the “dorm experiment” of applying GPUs to crack ImageNet competition. Today, Deep Learning is being applied across a variety of fields such as healthcare, genomics, drug discovery, retail, oil and gas manufacturing, and countless other industries and domains such as astrophysics and quantum computing.
However, as a Deep Learning Engineer, how aware are we of the core technical problems across these industries? How are these problems formulated to be solved in a typical framework of classification, detection, segmentation etc? What type of, what quality of, and what quantity of data is Deep Learning being applied upon?
As Deep Learning Engineers, we are most often exposed to text, photo, or video datasets. However, how much do we know about formulating and solving say a drug discovery problem?
This talk will take a deep dive into data and formulations in traditional (e.g., retail) and non-traditional fields (e.g., genomics) to draw similarities and differences between different Deep Learning approaches. If you wish to discover new planetary systems, come, attend this talk.
Key Takeaways:
- Collection of type/Quality/Quantity of data in non-traditional domains such as genomics, astrophysics
- Extensions of well-known neural networks such as convolutional neural networks to handle multi-dimensional datasets
- Formulations, success stories, and the impact of deep learning in traditional and non-traditional domains
