O slideshow foi denunciado.
Utilizamos seu perfil e dados de atividades no LinkedIn para personalizar e exibir anúncios mais relevantes. Altere suas preferências de anúncios quando desejar.

Deep learning introduction

115 visualizações

Publicada em

Presentation on the Deep Learning Introduction

Publicada em: Tecnologia
  • Seja o primeiro a comentar

  • Seja a primeira pessoa a gostar disto

Deep learning introduction

  1. 1. DEEP LEARNING SHASHI JEEVAN M P
  2. 2. SPEAKER • M. Tech. from IIT Kharagpur • Inventor of US Patent # 6,609,084, Issue Date: August 19, 2003 • Two decades of experience in Software Industry • My Blog (https://shashijeevan.com)
  3. 3. DEFINITION • Artificial Intelligence • Machine Learning • Deep Learning
  4. 4. MACHINE LEARNING DEFINITION • Machine Learning: Field of study that gives computers the ability to learn without being explicitly programmed. ~Arthur Samuel, 1959 • Well posed Learning Problem: A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E. ~Tom Mitchell, 1998
  5. 5. MACHINE LEARNING TYPES • Supervised Learning – Naïve Bayes, SVM, Artificial Neural Nets, Random Forest • Unsupervised Learning – K Means Clustering • Reinforcement Learning – Model Free Learning, MDP, Q Learning • Semi Supervised Learning – GAN (New)
  6. 6. LINEAR REGRESSION • Best fit line
  7. 7. LINEAR REGRESSION Best fit line Y = m * X + B Minimize Errors using Least Squares method
  8. 8. APPLICATIONS • Netflix • Uber https://eng.uber.com/michelangelo/ • Amazon shopping
  9. 9. TECHNIQUES • Loss Function • Gradient Descent • Back propagation
  10. 10. NETWORK ARCHITECTURES • https://becominghuman.ai/cheat-sheets-for-ai-neural- networks-machine-learning-deep-learning-big-data- 678c51b4b463
  11. 11. BUILDING MODEL • Layers are organized – Convolutional, Recurrent • Activation – Sigmoid, ReLU, Softmax • Loss function • Optimizer
  12. 12. TRAINING • Batch • Epoch • Training Set/Testing Set • Loss • Accuracy
  13. 13. INFERENCE • Saved model is used • Input is processed and a prediction is made
  14. 14. MODEL EXCHANGE STANDARDS
  15. 15. DEMO • Training • Keras MNIST • Inference • http://myselph.de/neuralNet.html
  16. 16. RESOURCES • https://www.analyticsvidhya.com • https://www.coursera.org/ • https://machinelearningmastery.com/handwritten-digit- recognition-using-convolutional-neural-networks-python- keras/

×