4.6 Read 21 reviews
Overview
What you will study
The Practical Deep Neural Networks AI - Best Practices for Gradient Learning course at University Teknologi PETRONAS reviews common training loss functions and regularization strategies which improve the convergence of gradient learning.
With a good understanding of these fundamentals, we will study the motivation and implementation of input, weight and activation normalizations and clipping techniques that have been commonly used to stabilize gradient learning across multiple different network architectures.
We will discuss a numerical technique to check gradients to assess the success of gradient learning. Finally, we will study methods to enhance learning convergence through adaptive learning algorithms.
Programme Structure
The program focuses on:- Gradient descent and backpropagation learning
- Challenges of managing gradient learning
- Training hyperparameters
- Cost functions
- Cost function regularization strategies
- Weightage between data and regularized portions
- Gradient checking and gradient clipping
- Dropout regularization
- Weight initialization and normalization
- Activation Normalizations(Batch, Layer, Instance, Group, Scale)
- Input normalization and decorrelation
- Adaptive gradient learning
Key information
Duration
- Part-time
- 2 days
Start dates & application deadlines
- StartingApplication deadline not specified.
Language
English
Delivered
Online
Campus Location
- Seri Iskandar, Malaysia
Disciplines
Artificial IntelligenceWhat students do after studying
Join for free or log in to access our complete career info list.
Academic requirements
We are not aware of any specific GRE, GMAT or GPA grading score requirements for this programme.
English requirements
We are not aware of any English requirements for this programme.
Other requirements
General requirements
- Engineers and researchers from all industries who need to implement deep neural networks AI.
- Engineers, researchers and consultants who have difficulty improving the performance of their deep neural network AI systems for industry 4.0 Prerequisite: Participants should have some basic knowledge and hands-on experience with training and setting up a deep neural network.
Tuition Fees
Tuition fees are shown in and the most likely applicable fee is shown based on your nationality.
-
International Applies to you
Applies to youNon-residents1305 MYR / full≈ 1305 MYR / full -
Domestic Applies to you
Applies to youCitizens or residents1305 MYR / full≈ 1305 MYR / full
Additional Details
- Professionals: MYR1,450
Funding
Improve page content