Deep learning with TensorFlow Probability in cancer prediction with reporting confidence
Who is this presentation for?
- Deep learning practitioners, particularly those focused on developing deep learning models in a regulated industry
Deep learning, which involves powerful black box predictors, has achieved state-of-the-art performance in medical imaging analysis, such as segmentation and classification for diagnosis. However, in spite of these successes, these methods focus exclusively on improving accuracy of point predictions without assessing the quality of their outputs. Knowing how much confidence there is in a prediction is essential for gaining clinicians’ trust in the technology.
Biraja Ghoshal walks you through creating probabilistic image classification models using Bayesian deep learning with TensorFlow Probability. Learn how to classify images using deep learning, implement convolutional neural networks, improve the model by batch normalization, dropout, and estimate uncertainty in cancer prediction.
- Familiarity with deep learning
- Experience with image classification (useful but not required)
What you'll learn
- Understand the inherent complexity of practical deep learning solutions
- Explore applications of deep learning in radiology
- Learn how to gain trust in deep learning model prediction leveraging TensorFlow Probability
Tata Consultancy Service
Biraja Ghoshal is a computer consultant with Tata Consultancy Service. He has 21 years of software development, architecture, and systems engineering expertise in information management and mining massive datasets technologies. Biraja assists clients to apply analytic capabilities using big data platforms to improve performance and optimize decision making with high-quality, actionable insights. Biraja is also interested in machine learning, cognitive computing, and artificial intelligence topics.
For conference registration information and customer service
For more information on community discounts and trade opportunities with O’Reilly conferences
For information on exhibiting or sponsoring a conference
For media/analyst press inquires