The intent of this glossary is to provide clear definitions of the technical terms specific to deep artificial neural networks. It is a work in progress. An activation, or activation function, for a neural network is defined as the mapping of the input to the output via a non-linear transform function at each "node", which is simply a locus of computation within the net. Each layer in a neural net consists of many nodes, and the number of nodes in a layer is known as its width. Activation algorithms are the gates that determine, at each node in the net, whether and to what extent to transmit the signal the node has received from the previous layer. A combination of weights (coefficients) and biases work on the input data from the previous layer to determine whether that signal surpasses a given treshhold and is deemed significant. Those weights and biases are slowly updated as the neural net minimizes its error; i.e. the level of nodes' activation change in the course of learning. These activation functions allow neural networks to make complex boundary decisions for features at various levels of abstraction. Adadelta is an updater, or learning algorithm, related to gradient descent. Unlike SGD, which applies the same learning rate to all parameters of the network, Adadelta adapts the learning rate per parameter. Adagrad, short for adaptive gradient, is an updater or learning algorithm that adjust the learning rate for each parameter in the net by monitoring the squared gradients in the course of learning. It is a substitute for SGD, and can be useful when processing sparse data. Affine is a fancy word for a fully connected layer in a neural network. "Fully connected" means that all the nodes of one layer connect to all the nodes of the subsequent layer. A restricted Boltzmann machine, for example, is a fully connected layer. Convolutional networks use affine layers interspersed with both their namesake convolutional layers (which create feature maps based on convolutions) and downsampling layers, which throw out a lot of data and only keep the maximum value. "Affine" derives from the Latin affinis, which means bordering or connected with. Each connection, in an affine layer, is a passage whereby input is multiplied by a weight and added to a bias before it accumulates with all other inputs at a given node, the sum of which is then passed through an activation function: e.g.

Twitter has just announced it has picked up London-based Fabula AI. The deep learning startup has been developing technology to try to identify online disinformation by looking at patterns in how fake stuff vs genuine news spreads online -- making it an obvious fit for the rumor-riled social network. Social media giants remain under increasing political pressure to get a handle on online disinformation to ensure that manipulative messages don't, for example, get a free pass to fiddle with democratic processes. Twitter says the acquisition of Fabula will help it build out its internal machine learning capabilities -- writing that the UK startup's "world-class team of machine learning researchers" will feed an internal research group it's building out, led by Sandeep Pandey, its head of ML/AI engineering. This research group will focus on "a few key strategic areas such as natural language processing, reinforcement learning, ML ethics, recommendation systems, and graph deep learning" -- now with Fabula co-founder and chief scientist, Michael Bronstein, as a leading light within it.

Reinforcement learning is currently one of the hottest topics in machine learning. For a recent conference we attended (the awesome Data Festival in Munich), we've developed a reinforcement learning model that learns to play Super Mario Bros on NES so that visitors, that come to our booth, can compete against the agent in terms of level completion time. The promotion was a great success and people enjoyed the „human vs. machine" competition. There was only one contestant who was able to beat the AI by taking a secret shortcut, that the AI wasn't aware of. Also, developing the model in Python was a lot of fun. So, I decided to write a blog post about it that covers some of the fundamental concepts of reinforcement learning as well as the actual implementation of our Super Mario agent in TensorFlow (beware, I've used TensorFlow 1.13.1, TensorFlow 2.0 was not released at the time of writing this article). Most machine learning models have an explicit connection between inputs and outputs that does not change during training time. Therefore, it can be difficult to model or predict systems, where the inputs or targets themselves depend on previous predictions. However, often, the world around the model updates itself with every prediction made. What sounds quite abstract is actually a very common situation in the real world: autonomous driving, machine control, process automation etc. -- in many situations, decisions that are made by models have an impact on their surroundings and consequently on the next actions to be taken. Classical supervised learning approaches can only be used to a limited extend in such kinds of situations. To solve the latter, machine learning models are needed that are able to cope with time-dependent variation of inputs and outputs that are interdependent. This is where reinforcement learning comes into play. In reinforcement learning, the model (called agent) interacts with its environment by choosing from a set of possible actions (action space) in each state of the environment that cause either positive or negative rewards from the environment. Think of rewards as an abstract concept of signalizing that the action taken was good or bad. Thereby, the reward issued by the environment can be immediate or delayed into the future. By learning from the combination of environment states, actions and corresponding rewards (so called transitions), the agent tries to reach an optimal set of decision rules (the policy) that maximize the total reward gathered by the agent in each state. In reinforcement learning we often use a learning concept called Q-learning. Q-learning is based on so called Q-values, that help the agent determining the optimal action, given the current state of the environment. Q-values are „discounted" future rewards, that our agent collects during training by taking actions and moving through the different states of the environment.

The Global Financial Crisis (hereafter, GFC) of 2007–2008 had far reaching financial and legal consequences, affecting millions of livelihoods. Sparked by the proliferation of subprime mortgages and exemplified by fall of Lehman Brothers, it's aftershocks were widely felt around the world. Following a period of recovery and growth, the world plunged into the European Sovereign Debt Crisis (hereafter, ESDC) beginning in late 2009, the effects of which have been argued to be still ongoing today. As global markets tend to operate in a cyclical fashion, scenario planning for the next financial crisis is not a matter of if, but when. Indeed, a Google search would lead to hundreds of differing opinions on the matter, from conclusions based on quantitative metrics to others based on the prophecies of Nostrodamus.

Two of the greatest obstacles to getting started with today's deep learning systems have been the lack of truly "point and click" interfaces to creating new models and the immense complexities in scaling machine learning workflows to the production scales businesses work with. While there are tools that make creating new models more straightforward than writing reams of code, few permit the creation of truly production-grade systems built upon transfer learning from some of the world's largest training datasets and bleeding edge algorithms and automated tuning workflows. Similarly, building massive scalable robust machine learning pipelines can be done with many different tools today, but the process is far from a point-and-click experience or a few lines of code. Google's announcements over the last few years offer a glimpse into how the cloud is transforming the AI workflow experience. Google's approach to enabling one-click creation of state-of-the-art deep learning models can be seen through in its AutoML series of products.

One of the most intriguing of these is weather forecasting, which currently has an 80% accuracy rate for seven-day forecasts, but drops sharply to 50% by the 10th day out. Deep-learning AI like the Numerical Weather Prediction (NWP) or IBM's Watson are able to constantly analyze all of the data coming in from weather balloons, radar, satellites and old weather models, noticing and accounting for patterns and trends meteorologists would never have been able to spot on their own.

Spiking neural networks (SNNs) with a lattice architecture are introduced in this work, combining several desirable properties of SNNs and self-organized maps (SOMs). Networks are trained with biologically motivated, unsupervised learning rules to obtain a self-organized grid of filters via cooperative and competitive excitatory-inhibitory interactions. Several inhibition strategies are developed and tested, such as (i) incrementally increasing inhibition level over the course of network training, and (ii) switching the inhibition level from low to high (two-level) after an initial training segment. During the labeling phase, the spiking activity generated by data with known labels is used to assign neurons to categories of data, which are then used to evaluate the network's classification ability on a held-out set of test data. Several biologically plausible evaluation rules are proposed and compared, including a population-level confidence rating, and an $n$-gram inspired method. The effectiveness of the proposed self-organized learning mechanism is tested using the MNIST benchmark dataset, as well as using images produced by playing the Atari Breakout game.

Efforts to understand the generalization mystery in deep learning have led to the belief that gradient-based optimization induces a form of implicit regularization, a bias towards models of low "complexity." We study the implicit regularization of gradient descent over deep linear neural networks for matrix completion and sensing, a model referred to as deep matrix factorization. Our first finding, supported by theory and experiments, is that adding depth to a matrix factorization enhances an implicit tendency towards low-rank solutions, oftentimes leading to more accurate recovery. Secondly, we present theoretical and empirical arguments questioning a nascent view by which implicit regularization in matrix factorization can be captured using simple mathematical norms. Our results point to the possibility that the language of standard regularizers may not be rich enough to fully encompass the implicit regularization brought forth by gradient-based optimization.

Forming perceptual groups and individuating objects in visual scenes is an essential step towards visual intelligence. This ability is thought to arise in the brain from computations implemented by bottom-up, horizontal, and top-down connections between neurons. However, the relative contributions of these connections to perceptual grouping are poorly understood. We address this question by systematically evaluating neural network architectures featuring combinations of these connections on two synthetic visual tasks, which stress low-level "gestalt" vs. high-level object cues for perceptual grouping. We show that increasing the difficulty of either task strains learning for networks that rely solely on bottom-up processing. Horizontal connections resolve this limitation on tasks with gestalt cues by supporting incremental spatial propagation of activities, whereas top-down connections rescue learning on tasks featuring object cues by propagating coarse predictions about the position of the target object. Our findings disassociate the computational roles of bottom-up, horizontal and top-down connectivity, and demonstrate how a model featuring all of these interactions can more flexibly learn to form perceptual groups.

Deep Learning has attracted considerable attention across multiple application domains, including computer vision, signal processing and natural language processing. Although quite a few single node deep learning frameworks exist, such as tensorflow, pytorch and keras, we still lack a complete processing structure that can accommodate large scale data processing, version control, and deployment, all while staying agnostic of any specific single node framework. To bridge this gap, this paper proposes a new, higher level framework, i.e. Nemesyst, which uses databases along with model sequentialisation to allow processes to be fed unique and transformed data at the point of need. This facilitates near real-time application and makes models available for further training or use at any node that has access to the database simultaneously. Nemesyst is well suited as an application framework for internet of things aggregated control systems, deploying deep learning techniques to optimise individual machines in massive networks. To demonstrate this framework, we adopted a case study in a novel domain; deploying deep learning to optimise the high speed control of electrical power consumed by a massive internet of things network of retail refrigeration systems in proportion to load available on the UK National Grid (a demand side response). The case study demonstrated for the first time in such a setting how deep learning models, such as Recurrent Neural Networks (vanilla and Long-Short-Term Memory) and Generative Adversarial Networks paired with Nemesyst, achieve compelling performance, whilst still being malleable to future adjustments as both the data and requirements inevitably change over time.