Lecture 3 continues our discussion of linear classifiers. We introduce the idea of a loss function to quantify our unhappiness with a model's predictions, and discuss two commonly used loss functions for image classification: the multiclass SVM loss and the multinomial logistic regression loss. We introduce the idea of regularization as a mechanism to fight overfitting, with weight decay as a concrete example. We introduce the idea of optimization and the stochastic gradient descent algorithm. We also briefly discuss the use of feature representations in computer vision.

This project combines two of the recent architectures StackGAN and ProGAN for synthesizing faces from textual descriptions. The project uses Face2Text dataset which contains 400 facial images and textual captions for each of them. The data can be obtained by contacting either the RIVAL group or the authors of the aforementioned paper. The code is present in the implementation/ subdirectory. The implementation is done using the PyTorch framework.

Baidu Research today announced it has developed a deep learning algorithm that in initial tests outperforms human pathologists in its ability to identify breast cancer metastasis. The convolutional neural net was trained by splitting 400 large images into grids of tens of thousands of smaller images, then randomly selecting 200,000 of those smaller images. The algorithm then performs analysis to classify each of the smaller photos as well as its neighboring cells. A variety of algorithms have been introduced to help pathologists examine images that can be gigabytes in size by cutting them into smaller parts. Baidu Research's algorithm attempts to move this technique forward by mimicking a pathologist's method to examine the area surrounding a breast cancer tumor cell, at once examining individual cells and nearby cells.

Today the Apache MXNet community is announcing the 1.2.0 release of the Apache MXNet deep learning framework -- faster, easier to use: ONNX, MKL-DNN support, mixed precision training and many more features! Check the full release note here. MKL-DNN integration: MXNet now integrates with Intel MKL-DNN to accelerate neural network operators: Convolution, Deconvolution, FullyConnected, Pooling, Batch Normalization, Activation, LRN, Softmax, as well as some common operators like sum and concat. This integration allows NDArray to contain data with MKL-DNN layouts and reduces data layout conversion overhead to get the maximal performance from MKL-DNN. More details are available in this blog post.

Cardiac arrhythmia is a condition where heart beat is irregular. The goal of this paper is to apply deep learning techniques in the diagnosis of cardiac arrhythmia using ECG signals with minimal possible data pre-processing. We employ convolutional neural network (CNN), recurrent structures such as recurrent neural network (RNN), long short-term memory (LSTM) and gated recurrent unit (GRU) and hybrid of CNN and recurrent structures to automatically detect the abnormality. Unlike the conventional analysis methods, deep learning algorithms don’t have feature extraction based analysis methods. The optimal parameters for deep learning techniques are chosen by conducting various trails of experiments.

Let's take a deep dive into reinforcement learning. In this article, we will tackle a concrete problem with modern libraries such as TensorFlow, TensorBoard, Keras, and OpenAI gym. You will see how to implement one of the fundamental algorithms called deep $Q$-learning to learn its inner workings. Regarding the hardware, the whole code will work on a typical PC and use all found CPU cores (this is handled out of the box by TensorFlow). The problem is called Mountain Car: A car is on a one-dimensional track, positioned between two mountains. The goal is to drive up the mountain on the right (reaching the flag). However, the car's engine is not strong enough to climb the mountain in a single pass. Therefore, the only way to succeed is to drive back and forth to build up momentum. This problem was chosen because it is simple enough to find a solution with reinforcement learning in minutes on a single CPU core. However, it is complex enough to be a good representative.

For more content like this, follow Insight and Emmanuel on Twitter. An image is worth a thousand words, and even more lines of code. Many products fundamentally appeal to our perception. When browsing through outfits on clothing sites, looking for a vacation rental on Airbnb, or choosing a pet to adopt, the way something looks is often an important factor in our decision. The way we perceive things is a strong predictor of what kind of items we will like, and therefore a valuable quality to measure.

We are in the midst of an unprecedented surge of investment into artificial intelligence (AI) research and applications. Within that, discussions about'ethics' are taking centre stage to offset some of the potentially negative impacts of AI on society. Mona Sloane writes that to achieve a sustainable shift towards such fields, we need a more holistic approach to the relationship between technology, data, and society. In June 2018, the Mayor of London released a new report that identifies London's'unique strengths as a global hub of Artificial Intelligence' and positions the capital as'The AI Growth Capital of Europe'. This plea coincides with the government's focus on'AI & Data Economy' as the first out of four'Grand Challenges' to put the UK'at the forefront of the industries of the future'.

Police investigators have more sources for investigative leads than ever before, which is both a blessing and a curse. It helps that there is so much more information available, but identifying the relevant from the irrelevant data is time-consuming and frustrating. A concept called "deep learning" helps to filter out the noise and find the details investigators are looking for. Deep learning differs from artificial intelligence. Artificial intelligence can be useful for crime analysis, identifying connections between data points that are not readily apparent.

Researchers at the Allen Institute for Cell Science, a Seattle research group founded by Microsoft co-founder Paul Allen, have created the first predictive 3D model of a live human cell. Using the model, scientists can digitally visualize and even manipulate cell behavior on a computer screen. Called the Allen Integrated Cell, the model is the result of deep learning training with tens of thousands of high-quality cell images. It's able to identify subcellular structures and project a 3D, multilayered image of a cell that shows how all its components interact simultaneously -- something that has never been visualized in this way before. "To me, it's the closest thing I've ever seen in science to magic," said Rick Horwitz, Executive Director of the Allen Institute for Cell Science.