Corporate Deep Learning Fundamentals Training Course

What is a neural network?

1. Introduction to artificial neural networks
2. Structure and function of neurons
3. Layers in neural networks: Input, hidden, and output
4. Activation functions: Purpose and types

2. Why Deep Learning?

1. Comparison with traditional machine learning
2. Applications and success stories
3. Advantages of Deep Learning

3. How to choose between deep neural networks?

1. Factors to consider: Accuracy, training time, and complexity
2. Overview of network architectures

4. An old problem: The vanishing gradient

1. Understanding the vanishing gradient problem
2. Impacts on training deep neural networks
3. Solutions and workarounds

5. Restricted Boltzmann machines

1. Introduction and structure
2. Applications in feature learning and dimensionality reduction
3. Training restricted Boltzmann machines

6. Deep belief networks

1. Understanding deep belief networks
2. Stacked restricted Boltzmann machines
3. Use cases and training methods

7. Convolutional networks

1. Basics of convolutional neural networks (CNNs)
2. Key components: convolutional layers, pooling layers
3. Applications in image and video recognition<

8. Recurrent nets

1. Introduction to recurrent neural networks (RNNs)
2. Long short-term memory (LSTM) and gated recurrent units (GRUs)
3. Applications in time series analysis and natural language processing

9. Autoencoders

1. Understanding autoencoders and their structure
2. Variants: denoising autoencoders, sparse autoencoders
3. Applications in data compression and noise reduction

10. Recursive neural tensor nets

1. Basics and architecture
2. Applications in natural language processing and semantic analysis
3. Differences between recursive and recurrent neural networks

What is a Deep Learning platform?

1. Definition and key features
2. Benefits over traditional development environments

2. H2O.ai

1. Overview and key features
2. Machine learning algorithms supported

3. Dato GraphLab

1. Introduction to GraphLab
2. Key capabilities and features

What is a deep learning library?

1. Role in simplifying deep learning development
2. Criteria for choosing a deep learning library

2. Theano

1. Core features and strengths
2. Building blocks: Symbolic expressions, compilers
3. Integration with other libraries and frameworks

3. Deeplearning4j

1. Overview and ecosystem
2. Key features: Distributed deep learning, scalability
3. Cross-platform compatibility 

4. Torch

1. Introduction to Torch and its design philosophy
2. Key libraries and their functionalities
3. Community contributions and extensions

5. Caffe

1. Basics of Caffe and its user-friendly approach
2. Model zoo: Pre-trained models availability