tensorflowtutorial
tensorflowtutorial
TensorFlow
747 posts
        TensorFlow, Pyhton, Deep Learning, Developer      
Don't wanna be here? Send us removal request.
tensorflowtutorial · 7 years ago
Photo
Tumblr media
To use a trivial example, don't use convolution layers on structured data, as you will be ignoring many important correlations across features that are not spatially close in your original encoding of the data, but that are close in information space.
20 notes · View notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
Making assumptions about the structure of correlations in the data will *reduce* the expressiveness of the model (that's the point: dramatically reduce the search space to make search more data-efficient and faster). Which is why it is important to make *correct* assumptions. https://t.co/2g6ml5W68A
7 notes · View notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
When we find ourselves in a conflict, it’s a good idea to pause to acknowledge our shared values and our shared goals, and remind ourselves that we are, almost certainly, on the same side.
7 notes · View notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
(Which is naturally the case past the first layer of any convnet.)
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
For instance, convolution is preferable to dense layers for data dimensions that are locally autocorrelated and translation-invariant. And depthwise separable convolutions are superior to convolution for any data that is locally autocorrelated and where channels are decorrelated.
17 notes · View notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
A good deep learning architecture is one that introduces correct priors about the problem at hand -- in particular, about the structure of the correlations found in the data.
1 note · View note
tensorflowtutorial · 7 years ago
Photo
Tumblr media
In particular, if you understand something clearly, you should be able to describe it in precise algorithmic terms to a computer: you should be able to implement it from scratch (as a simulation, as a framework, etc).
2 notes · View notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
A popular quote goes "if you can't explain it in simple terms, you don't understand it well enough" (often incorrectly attributed to Einstein or Feynman). I think a more accurate take is: "if you can't explain it in arbitrarily precise terms, you don't understand it well enough"
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
@Hannesvdm1974 Please send me a note (francois.chollet, that's gmail), I don't use Slack anymore
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
Affective computing aims to model how emotion impacts decision-making. @lmoroney talks with Christian Ramsey about using TensorFlow to model complex, emotional behavior and adding the affective layer to deep learning. Watch #TensorFlowMeets here → https://t.co/IQ1JDIAaIv https://t.co/BaZI5DibF8
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
Trust, good faith, and a cooperation mindset are the foundational layer of all relationships. Without them, you cannot have a society, or a company, etc. We should place these first, far above self-interest, since the loss of these spells the loss of everything else.
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
Can differentiable programming learn arbitrarily complex behaviors, up to superhuman performance, given a dense sampling of the behavior manifold (e.g. an infinite training data generator, such as self-play)? Yes. We knew that. Arguably AlphaGo was the concrete proof-of-concept.
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
Learn how to train your first deep neural network, using @GoogleColab to execute the code directly from the browser. @MagnusHyttsten shows how in the latest #CodingTensorFlow Watch here → https://t.co/wpE5hQVANZ https://t.co/QpRoA2C2wD
3 notes · View notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
This got more of a reaction than I expected, the need for abstraction in AI seems to be an increasingly popular topic. More on this from July 2017: https://t.co/9upBSqsOOb https://t.co/S5sVIg81p5
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
RT @Max_Fisher: What if Facebook’s effects are broader and more pernicious than hate speech or fake news? What if it scrambles the way entire communities determine right-or-wrong? What if it could be linked, by hard empirical data, to a national epidemic of violence? https://t.co/jSHzsN68IU
2 notes · View notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
@AuerbachKeller We need mathematical formalism for abstraction in AI & computer science in general (not category theory), which we will have "soon" (on the scale of scientific progress)
0 notes
tensorflowtutorial · 7 years ago
Photo
Tumblr media
🦒
1 note · View note