Wunda Deeplearning.ai last Lesson online, next time how many years? _ Wunda

Hi da pu ben. Today, the last lesson of Wunda's in-depth study series is online.

Last June, Wunda announced the Deeplearning.ai Entrepreneurship Project, which was unveiled in August: a series of 5-course in-depth learning courses--deep Learning specialization, designed to promote universal knowledge of deep learning.

The first 3 courses were launched on the line, but it was not until November that the 4th course came late, and after that, Miss Wu was plunged into a quiet period and started working on two other entrepreneurial projects: Landing.ai and Aifund. Until today (February 1), Deep Learning Specialization's 5th lesson was finally online, which is also the Deep Learning specialization of the last lesson.

The following is a brief introduction to the 5th lesson:

▌ Course Introduction

This course will teach you how to build a model of natural language, audio, and other sequence data. Based on depth learning, sequence algorithm has a great leap forward two years ago, and spawned many applications in the fields of speech recognition, music synthesis, chat robot, machine translation and natural language understanding.

Through this course, you will: understand how to build and train a cyclic neural network (RNN), and some widely used variants, such as GRU and lstm, that can apply sequence models to natural language problems, including literal synthesis. The ability to apply sequence models to audio applications, including speech recognition and music synthesis.

Applicable crowd:

Having completed the first, second and fourth classes, we also recommend the third lesson, which has a deep understanding of neural networks (including convolution networks) and wants to learn how to develop a circular neural network for learners.

Lecturer: Wunda, Kian Katanforoosh, Younes Bensouda Mourri

▌ Course Arrangement

Cycle sequence model of the first week

This week's knowledge points are circular neural networks. This type of model has been shown to perform well on time data, with several variants, including LSTM, GRU and bi-directional neural networks, which are included in this week's course.

The second week natural language processing and word embedding

Natural language processing and depth learning are particularly important combinations. Using the word vector representation and embedding layer, we can train the cyclic neural networks that perform well in various industries. Examples of applications include emotion analysis, object recognition, and machine translation.

Third-week sequence model and attention mechanism

The attention mechanism can enhance the sequence model. This algorithm will help your model understand where it should focus when giving a series of inputs. This week, you will also learn about speech recognition and how to handle audio data.

In the 2011, Wunda's machine learning legendary course was launched as a classic introduction to machine learning. 6 years later, Wunda from Baidu finally regained the status of "teacher", the introduction of a new curriculum--deep Learning specialization. However, Wunda, who is busy with many entrepreneurial projects, is clearly at a time when we will have to wait until the next time.