These programmers taught an AI how to understand tattoos

These programmers taught an #AI how to understand tattoos

  • The process involved using a deep-learning framework called Caffe, and feeding it data-sets with images representative of different tattoo styles.
  • Once the initial training session was complete, the AI could identify the style of a tattoo with pretty impressive accuracy.
  • While the AI isn’t implemented into the app yet (they’re still feeding it data) they intend to finish training the AI then go forward from there.
  • AI will help us to classify the remaining 250k pictures… Classification is really important for us because, based on it, we can show users personalized feeds depending on what styles they like, what artists they follow, what those artists are specialized in, etc.
  • Without AI to sort images, a person has to view each one, decide what style it represents, tag the image, and then create hashtags so that other users can find it.

A pair of developers at Tattoodo developed a neural-network capable of processing pictures of tattoos and determining the style of the ink in the images.
Continue reading “These programmers taught an AI how to understand tattoos”

TensorFlow Dev Summit 2017: Integrating Keras and TensorFlow

TensorFlow Dev Summit 2017: Integrating #Keras and #TensorFlow

  • My choice has fallen to a presentation by François Chollet of the deep learning library API Keras and its integration with TensorFlow.
  • As Dr. Chollet explains, Keras integrated with TensorFlow promises to streamline deep learning frameworks in ways that will be increasingly user-friendly, rendering the mass adoption of these software developments a more feasible reality:

    Dr. François Chollet is the primary author of Keras, developing this tool while at Research at Google.

  • For instance the way video with text data is processed with the Keras-TensorFlow integration is nicely described with the stack of CNNs, LSTMs and dense final layers with softmax being features explained by Dr. Chollet.
  • The best practises advised by Dr. Chollet about the initialization of recurrent weighs of  the neural network is worth to listen, even if the experienced practitioner feels bored.
  • A final note to the confirmation by Dr. Chollet of the capacity of TensorFlow to streamline  a CloudML or a hyperparameter tuning process with just a few lines of code, enabling a distributed training platform able to enhance big data computes with productivity gains.

I am briefly sharing a video from the last TensorFlow Dev Summit in February 2017. My choice has fallen to a presentation by François Chollet of the deep learning library API Keras and its integration with TensorFlow. As Dr. Chollet explains, Keras integrated with TensorFlow promises to streamline deep learning frameworks in ways that will…
Continue reading “TensorFlow Dev Summit 2017: Integrating Keras and TensorFlow”

China Gears Up in Artificial-Intelligence Race

#China Gears Up in Artificial-Intelligence Race, Baidu and others #AI #BigData @WSJ $

  • China Gears Up in Artificial Intelligence
  • China Gears Up in Artificial-Intelligence Race
  • Why Venture Capitalists Are Betting on Marketing Tech Over Ad Tech
  • China Sets New Tone in Drafting Cybersecurity Rules
  • Chinese technology companies, including Baidu, invest heavily in AI efforts

Read the full article, click here.


@cybersecboardrm: “#China Gears Up in Artificial-Intelligence Race, Baidu and others #AI #BigData @WSJ $”


China Circuit Columnist Li Yuan writes that the biggest buzz in China’s internet industry is about competing head-to-head with the U.S. and other tech powerhouses in artificial intelligence.


China Gears Up in Artificial-Intelligence Race