Conditional Image Generation with PixelCNN Decoders – slides

Awhile ago I presented and attempted to explain this work to our reading group:

van den Oord, A., Kalchbrenner, N., Vinyals, O., Espeholt, L., Graves, A., & Kavukcuoglu, K. (2016). Conditional Image Generation with PixelCNN Decoders. In D. D. Lee, M. Sugiyama, U. V Luxburg, I. Guyon, & R. Garnett (Eds.), NIPS (pp. 4790–4798). Retrieved from

And also dived a bit into their previous work,
van den Oord, A., Kalchbrenner, N., & Kavukcuoglu, K. (2016). Pixel Recurrent Neural Networks. Arxiv, 48. Retrieved from

While I usually post slides to the web shortly after, this time I’ve been scared to do so. There are a few critical points from this paper that I still don’t understand. And while I told myself that I would spend some time to figure this out, it is now months later, and I’ve taken no action. So as now is always the time to continue on in spite of the fear, I’ll let you, dear Internet, have these slides in all there erroneous ways.

Continue reading “Conditional Image Generation with PixelCNN Decoders – slides”

Convolutional Neural Networks for Adjacency Matrices

We had our work, BrainNetCNN, published in NeuroImage awhile ago,

Kawahara, J., Brown, C. J., Miller, S. P., Booth, B. G., Chau, V., Grunau, R. E., Zwicker, J., G., Hamarneh, G. (2017). BrainNetCNN: Convolutional neural networks for brain networks; towards predicting neurodevelopment. NeuroImage, 146(Feb), 1038–1049.

and I’ve meant to do a blog writeup about this. We recently released our code for BrainNetCNN on GitHub (based on Caffe), which implements the proposed filters designed for adjacency matrices.

We called this library Ann4Brains. In hindsight, we could have called this something more general and cumbersome like Ann4AdjacencyMatrcies, but I still like the zombie feel that Ann4Brains has.

We designed BrainNetCNN specifically with brain connectome data in mind. Thus the tag line of,

“Convolutional Neural Networks for Brain Networks”

seemed appropriate. However, after receiving some emails about using BrainNetCNN for other types of (non-connectome) data, I’ll emphasize that this approach can be applied to any sort of adjacency matrix, and not just brain connectomes.

The core contribution of this work is the filters designed for adjacency matrices themselves. So we’ll go through each of them. But first, let’s make sure we are clear on what the brain connectome (or adjacency matrix) is.

Continue reading “Convolutional Neural Networks for Adjacency Matrices”

Mastering the Game of Go – slides [paper explained]

This week I presented to our weekly reading group, this work:

Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., … Hassabis, D. (2016). Mastering the game of Go with deep neural networks and tree search. Nature, 529(7587), 484–489.

To quickly summarize this work…

Basically, they create a policy network, which is a convolutional neural network, that predicts the next move a human player would do from a board state. They create a value network, also a convolutional neural network, that predicts the outcome (win or lose) of the game given the current board state.
Continue reading “Mastering the Game of Go – slides [paper explained]”

TensorFlow – failed call to cuInit: CUDA_ERROR_UNKNOWN

Scenario: You’re trying to get your GPU to work in TensorFlow on a Ubuntu Laptop. You’ve already installed Tensorflow, Cuda, and Nvidia drivers.

You run python and import TensorFlow:

import tensorflow as tf

And you see encouraging messages like: "successfully opened CUDA library locally"

But in Python, when you run,


You get this cryptic error:

failed call to cuInit: CUDA_ERROR_UNKNOWN

Here’s how to fix this.
Continue reading “TensorFlow – failed call to cuInit: CUDA_ERROR_UNKNOWN”

caffe – Check failed: proto.SerializeToOstream(&output)

You suddenly get this error when training/saving a model in Caffe or saving a model in pycaffe.

io.cpp:69] Check failed: proto.SerializeToOstream(&output)
*** Check failure stack trace: ***

Here are two possible reasons for this error

  1. The directory the snapshot is trying to write the .caffemodel into does not exist
  2. You are out of disk space

Continue reading “caffe – Check failed: proto.SerializeToOstream(&output)”

multi-resolution-tract CNN with hybrid pretrained and skin-lesion trained layers

Our paper entitled: “Multi-resolution-Tract CNN with Hybrid Pretrained and Skin-Lesion Trained Layers” was accepted and presented as an oral talk in the Machine Learning in Medical Imaging (MLMI) Workshop (part of the MICCAI conference).

In this work, we used a convolutional neural network (CNN) to classify 10 different types of skin lesions, including melanoma and non-melanoma types.

The key technical contribution was to use multiple tracts (or paths) within the neural network, to train (and test) the network on an image using multiple image resolutions simultaneously. Additionally, we extended a CNN pretrained on a single image resolution to work for multiple image resolutions.

Here are our slides presented at MLMI (thanks Aïcha!) showing our deep learning approach to classify skin disease:
Continue reading “multi-resolution-tract CNN with hybrid pretrained and skin-lesion trained layers”

(deep convolutional) generative adversarial nets – slides

There’s this really neat new idea on how to train neural networks that recently came out know as generative adversarial nets (GAN).

The basic idea of a GAN is to train two networks to compete with each other (hence the name “adversarial“). One network (called the generator) creates images that look just like real images. The other network (called the discriminator) distinguishes between real images and those images the generator produced.

Thus the two networks compete with each other, where the generator generates images to fool the discriminator, and the discriminator discriminates between the generator’s images and real images.

Continue reading “(deep convolutional) generative adversarial nets – slides”

Deep Dreams and a Neural Algorithm of Artistic Style – slides and explanations

Perhaps you saw an earlier post I wrote about deep dreaming Prague pictures, and you said to your self, “self, I wish I knew more about the techniques to make those crazy looking pictures.”

Well you are in luck since I’ve now posted the slides where I attempted to explain these two works to our reading group: 1) Google’s DeepDream, and 2) A Neural Algorithm of Artistic Style.

Continue reading “Deep Dreams and a Neural Algorithm of Artistic Style – slides and explanations”

What is the derivative of ReLU?

Prague travel pictures and deep dreaming

Here is a combined short summary on my travels to the city of Prague in the Czech Republic along with corresponding images created using Google’s DeepDreams.

What is this DeepDreams you speak of?

Basically, DeepDream is a deep neural network that was trained to recognize objects from millions of images. A deep neural network is composed of a stack of layers. Basically, these layers learn image filters that when applied to an image classify the image (e.g., is this an image of a cat or a dog?).

You give DeepDream an image and specify a layer in the neural network. The original image is then slightly perturbed to create a modified image that causes the specified layer in the neural network to be more activated.

Early layers in the neural network are sensitive to low level concepts like the edges and textures in the image. So if you specify an early layer, your image will be modified to have edges and textures that most activate the early selected layer.

Example CNN. Image goes as input. Early layers (purple) are sensitive to things like edges and textures. Later layers (red) are sensitive to higher level concepts like faces.
Example CNN. Image goes as input. Early layers (purple) are sensitive to things like edges and textures. Later layers (red) are sensitive to higher level concepts like faces.

Later (or deeper) layers in the neural network are activated when they see higher level concepts such as faces. So any areas in the original image that slightly look like a face, will be modified to look more like a face.

Okay, but now you might ask, but what about Prague? How was your trip? Did you like the city?

Yeah it was nice! Thanks for asking. Did you want to see some pictures? Here’s one of an old building.

Original image of some building in Prague... can't remember what/if the significance of this picture was.
Original image of some building in Prague… can’t remember what the significance of this picture was.

Let’s try some deep dreaming on this. We’ll use the neural network known as VGG16 (it’s a famous neural network that performs very well on competitions). We’ll start by telling VGG16 (the neural network) to modify this image so that one of it’s middle layers becomes more activated. Specifically, we will activate layer conv3_1 from VGG16 (if you don’t know what conv3_1 means, that’s okay – it’s just a technical detail specifying what layer to use). This gives us this:

Same Prague building but using DeepDreams with VGG16 conv3_1
Same Prague building but using DeepDreams with VGG16 conv3_1

Now if we activate a deeper layer, conv5_2, we get this crazy looking image,

Continue reading “Prague travel pictures and deep dreaming”