TensorFlow or Keras? Which one should I learn? – Imploding Gradients – Medium

#TensorFlow or #Keras? Which one should I learn?

  • With plenty of libraries out there for deep learning, one thing that confuses a beginner in this field the most is which library to choose.Deep Learning libraries/frameworks as per popularity(Source : Google)In this blog post, I am only going to focus on Tensorflow and Keras.
  • And if Keras is more user-friendly, why should I ever use TF for building deep learning models?
  • You can tweak TF much more as compared to Keras.FunctionalityAlthough Keras provides all the general purpose functionalities for building Deep learning models, it doesn’t provide as much as TF.
  • Absolutely, check the example below:Playing with gradients in TensorFlow (Credits : CS 20SI: TensorFlow for Deep Learning Research)Conclusion (TL;DR)if you are not doing some research purpose work or developing some special kind of neural network, then go for Keras (trust me, I am a Keras fan!!)
  • But as we all know that Keras is going to be integrated in TF, it is wiser to build your network using tf.contrib.Keras and insert anything you want in the network using pure TensorFlow.

Deep learning is everywhere. 2016 was the year where we saw some huge advancements in the field of Deep Learning and 2017 is all set to see many more advanced use cases. With plenty of libraries out…
Continue reading “TensorFlow or Keras? Which one should I learn? – Imploding Gradients – Medium”

Google’s AI Guru Says That Great Artificial Intelligence Must Build on Neuroscience

Google’s #AI guru says that great artificial intelligence must build on neuro science

  • Demis Hassabis knows a thing or two about artificial intelligence: he founded the London-based AI startup DeepMind, which was purchased by Google for $650 million back in 2014.
  • In a paper published today in the journal Neuron, Hassabis and three coauthors argue that only by better understanding human intelligence can we hope to push the boundaries of what artificial intellects can achieve.
  • But it also points out that more recent advances haven’t leaned on biology as effectively, and that a general intelligence will need more human-like characteristics—such as an intuitive understanding of the real world and more efficient ways of learning.
  • As Hassabis explains in an interview with the Verge, artificial intelligence and neuroscience have become “two very, very large fields that are steeped in their own traditions,” which makes it “quite difficult to be expert in even one of those fields, let alone expert enough in both that you can translate and find connections between them.”
  • (Read more: Neuron, The Verge, “Google’s Intelligence Designer,” “Can This Man Make AI More Human?”)

Inquisitiveness and imagination will be hard to create any other way.
Continue reading “Google’s AI Guru Says That Great Artificial Intelligence Must Build on Neuroscience”

Hashtag Artificial Intelligence – Katherine Bailey – Medium

Good read

Hashtag #ArtificialIntelligence 

 #fintech @katherinebailey @Acquia #AI

  • Here’s one that was presented recently at a talk by one of Amazon’s top Machine Learning people:Artificial Intelligence: A system or service which can perform tasks that usually require human intelligenceThis is a fairly common way to define it.
  • Here’s a similar formulation from Nathan Benaich in his post 6 areas of AI and machine learning to watch closely:The ultimate goal of AI […] is to build machines capable of performing tasks and cognitive functions that are otherwise only within the scope of human intelligence.One problem with this definition is that it means the state of being an instance of Artificial Intelligence is temporary.
  • Just look at all the “What is the difference between Artificial Intelligence and Machine Learning?”
  • Really, why should Machine Learning be defined in relation to AI?The slippery definition issue above can be looked at as follows: it is the term “Artificial Intelligence” looking for things to refer to.
  • This is no more true now than it was 50 years ago but many smart people are utterly convinced of it.The term “Artificial Intelligence” has been around since the early days of computer science, when “thinking machines” were seen as the natural next step after programming basic logic.

The internet is awash with stories about something called Artificial Intelligence. Confusion around what it is is prompting many to proffer definitions of it, or corrections of wrong definitions…
Continue reading “Hashtag Artificial Intelligence – Katherine Bailey – Medium”

WHEN COMPUTERS LEARN TO SWEAR: – Jigsaw – Medium

Introducing Perspective, using machine learning to improve discussions online.

  • We think technology can help.Today, Google and Jigsaw are launching Perspective, an early-stage technology that uses machine learning to help identify toxic comments.
  • Through an API, publishers — including members of the Digital News Initiative — and platforms can access this technology and use it for their sites.HOW IT WORKSPerspective reviews comments and scores them based on how similar they are to comments people said were “toxic” or likely to make someone leave a conversation.
  • Each time Perspective finds new examples of potentially toxic comments, or is provided with corrections from users, it can get better at scoring future comments.Publishers can choose what they want to do with the information they get from Perspective.
  • Publishers could even just allow readers to sort comments by toxicity themselves, making it easier to find great discussions hidden under toxic ones.We’ve been testing a version of this technology with The New York Times, where an entire team sifts through and moderates each comment before it’s posted — reviewing up to 11,000 comments every day.
  • We’ve worked together to train models that allows Times moderators to sort through comments more quickly, and we’ll work with them to enable comments on more articles every day.WHERE WE GO FROM HEREPerspective joins the TensorFlow library and the Cloud Machine Learning Platform as one of many new machine learning resources Google has made available to developers.

Imagine trying to have a conversation with your friends about the news you read this morning, but every time you said something, someone shouted in your face, called you a nasty name or accused you…
Continue reading “WHEN COMPUTERS LEARN TO SWEAR: – Jigsaw – Medium”

When computers learn to swear: Using machine learning for better online conversations

Neat, Google trained a computer to understand whether a comment is toxic or not

  • Unfortunately, this happens all too frequently online as people try to discuss ideas on their favorite news sites but instead get bombarded with toxic comments.
  • According to the same report, online harassment has affected the lives of roughly 140 million people in the U.S., and many more elsewhere.
  • News organizations want to encourage engagement and discussion around their content, but find that sorting through millions of comments to find those that are trolling or abusive takes a lot of money, labor, and time.
  • As a result, many sites have shut down comments altogether.
  • Through an API, publishers—including members of the Digital News Initiative—and platforms can access this technology and use it for their sites.

Google and Jigsaw announce the launch of Perspective, an early-stage technology that uses machine learning to identify toxic comments.
Continue reading “When computers learn to swear: Using machine learning for better online conversations”

Design Patterns for Deep Learning Architectures

Design Patterns for #DeepLearning Architectures:  #BigData #DataScience #MachineLearning

  • We purposely use “pattern language” to reflect that the field of Deep Learning is a nascent, but rapidly evolving, field that is not as mature as other topics in computer science.
  • Each pattern describes a problem and offers alternative solutions.
  • You can find more details on this book at: A Pattern Language for Deep Learning .
  • Pattern Languages are languages derived from entities called patterns that when combined form solutions to complex problems.
  • Or you can check for updates at Design Patterns for Deep Learning

Deep Learning can be described as a new machine learning toolkit that has a high likelihood to lead to more advanced forms of artificial intelligence. The evidence for this is in the sheer number of breakthroughs that had occurred since the beginning of this decade. There is a new found optimism in the air and we are now again in a new AI spring. Unfortunately, the current state of deep learning appears to many ways to be akin to alchemy. Everybody seems to have their own black-magic methods of designing architectures. The field thus needs to move forward and strive towards chemistry, or perhaps even a periodic table for deep learning. Although deep learning is still in its early infancy of development, this book strives towards some kind of unification of the ideas in deep learning. It leverages a method of description called pattern languages.
Continue reading “Design Patterns for Deep Learning Architectures”

Deep Learning Machine Teaches Itself Chess in 72 Hours, Plays at International Master Level

#Deeplearning machine teaches itself chess in 72 hours, plays at international master level:

  • But while computers have become faster, the way chess engines work has not changed.
  • Lai says Giraffe takes about 10 times longer than a conventional chess engine to search the same number of positions.
  • Deep Learning Machine Teaches Itself Chess in 72 Hours, Plays at International Master Level
  • Lai says that matches the best chess engines in the world.
  • Lai has created an artificial intelligence machine called Giraffe that has taught itself to play chess by evaluating positions much more like humans and in an entirely different way to conventional chess engines.

In a world first, a machine plays chess by evaluating the board rather than using brute force to work out every possible move.
Continue reading “Deep Learning Machine Teaches Itself Chess in 72 Hours, Plays at International Master Level”

IT-Trends 2016 for the insurance industry

.@MunichRe's (insurance) industry #IT-trends 2016
(src: ):
#BigData #IoT #IoE #AI #ML #SmartX

  • The publication is available exclusively to Munich Re clients.
  • The objective of the IT Trend Radar 2016 is to identify relevant new technologies for ERGO, Munich Re and MEAG and evaluate them from a group perspective.
  • Those who identify the potential of new technologies, trends and technological progress at an early stage have a competitive advantage and build the foundation for innovation.
  • Topics Online RSS feed Recommend page by e-mail Contact
  • We focus on the potential for innovation offered by individual trends, and review their suitability in practice for the reinsurance and primary insurance sector.

Read the full article, click here.


@DiegoKuonen: “.@MunichRe’s (insurance) industry #IT-trends 2016
(src: ):
#BigData #IoT #IoE #AI #ML #SmartX”


Those who identify the potential of new technologies, trends and technological progress at an early stage have a competitive advantage and build the foundation for innovation. Constantly increasing technological complexity is leading to a growing number of areas that are relevant for our company to keep an eye on.


IT-Trends 2016 for the insurance industry

Design Patterns for Deep Learning Architectures

Design Patterns for #DeepLearning #Architectures. #BigData #MachineLearning #DataScience #AI

  • We purposely use “pattern language” to reflect that the field of Deep Learning is a nascent, but rapidly evolving, field that is not as mature as other topics in computer science.
  • Each pattern describes a problem and offers alternative solutions.
  • You can find more details on this book at: A Pattern Language for Deep Learning
  • Pattern Languages are languages derived from entities called patterns that when combined form solutions to complex problems.
  • There are patterns that we describe that are not actually patterns, but rather may be fundamental concepts.

Read the full article, click here.


@gp_pulipaka: “Design Patterns for #DeepLearning #Architectures. #BigData #MachineLearning #DataScience #AI”


Deep Learning can be described as a new machine learning toolkit that has a high likelihood to lead to more advanced forms of artificial intelligence. The evidence for this is in the sheer number of breakthroughs that had occurred since the beginning of this decade. There is a new found optimism in the air and we are now again in a new AI spring. Unfortunately, the current state of deep learning appears to many ways to be akin to alchemy. Everybody seems to have their own black-magic methods of designing architectures. The field thus needs to move forward and strive towards chemistry, or perhaps even a periodic table for deep learning. Although deep learning is still in its early infancy of development, this book strives towards some kind of unification of the ideas in deep learning. It leverages a method of description called pattern languages.


Design Patterns for Deep Learning Architectures