Google Launches Deep Learning with TensorFlow MOOC
Google and Udacity have partnered for a new self-paced course on deep learning and TensorFlow, starting immediately.
Deep learning is still everywhere, and its prevalence is steadily growing. Much of the current excitement surrounding machine learning is deep neural network-based, and practitioners and researchers alike are attracted to the subfield for its perceived importance, its ability to tackle interesting challenges, and its promise of valuable results. The current wave of deep learning research marks a major event in the history of machine learning and artificial intelligence.
Machine learning powerhouse Google released TensorFlow, its internal deep learning library, into the wild last November, and it has achieved quick success and accolades from those in industry and research alike. While no reliable statistics on deep learning library marketshare are available (or really even possible to any degree of accuracy), TensorFlow is already on the shortlist of go-to options in the space, and its backing by Google may lead to de facto status at some point in the future.
With interest in both deep learning and TensorFlow reaching remarkable heights, Google has just announced a new Deep Learning Course developed in partnership with Udacity.
The course is made up of four lectures, covering the following:
Lecture 1 - Focuses on machine learning basics, including setting up data and experiments, and training simple classification models.
Lecture 2 - Builds on the fundamentals, exploring how to make models deeper, and exploring scalability issues such as hyperparameter tuning and regularization.
Lecture 3 - All about convolutional networks and image recognition.
Lecture 4 - Explores models for text and sequences in general, with embeddings and recurrent neural networks.
Google states that its aim in designing the course is to provide machine learning enthusiasts a concise pathway to quickly become familiar with deep learning. It's also a clever move from both marketing and engineering standpoints: the more practitioners that get their hands on TensorFlow, the better it is for both TensorFlow's market share and the number of eyes on, and potentially improving, its code base.
But it really comes down to this: getting instruction on using the hottest deep learning library from its creators and top-tier deep learning specialists is a huge positive, and anyone interested in deep learning or getting a handle on TensorFlow would do well to check out this (free) self-paced course. Vincent Vanhoucke, a Principal Scientist at Google, is a technical lead and manager in Google’s deep learning infrastructure team, and is the course's lead instructor.
Related:
Machine learning powerhouse Google released TensorFlow, its internal deep learning library, into the wild last November, and it has achieved quick success and accolades from those in industry and research alike. While no reliable statistics on deep learning library marketshare are available (or really even possible to any degree of accuracy), TensorFlow is already on the shortlist of go-to options in the space, and its backing by Google may lead to de facto status at some point in the future.
With interest in both deep learning and TensorFlow reaching remarkable heights, Google has just announced a new Deep Learning Course developed in partnership with Udacity.
The course is made up of four lectures, covering the following:
Lecture 1 - Focuses on machine learning basics, including setting up data and experiments, and training simple classification models.
Lecture 2 - Builds on the fundamentals, exploring how to make models deeper, and exploring scalability issues such as hyperparameter tuning and regularization.
Lecture 3 - All about convolutional networks and image recognition.
Lecture 4 - Explores models for text and sequences in general, with embeddings and recurrent neural networks.
Google states that its aim in designing the course is to provide machine learning enthusiasts a concise pathway to quickly become familiar with deep learning. It's also a clever move from both marketing and engineering standpoints: the more practitioners that get their hands on TensorFlow, the better it is for both TensorFlow's market share and the number of eyes on, and potentially improving, its code base.
But it really comes down to this: getting instruction on using the hottest deep learning library from its creators and top-tier deep learning specialists is a huge positive, and anyone interested in deep learning or getting a handle on TensorFlow would do well to check out this (free) self-paced course. Vincent Vanhoucke, a Principal Scientist at Google, is a technical lead and manager in Google’s deep learning infrastructure team, and is the course's lead instructor.
Related: