Machine Learning seminars

Deep neural networks (DNNs) with the flexibility to learn good top-layer representations have eclipsed shallow kernel methods without that flexibility. Here, we take inspiration from deep neural networks to develop a new family of deep kernel method. In a deep kernel method, there is a kernel at every layer, and the kernels are jointly optimized to improve performance (with strong regularisation). We establish the representational power of deep kernel methods, by showing that they perform exact inference in an infinitely wide Bayesian neural network or deep Gaussian process. Next, we conjecture that the deep kernel machine objective is unimodal, and give a proof of unimodality for linear kernels. Finally, we exploit the simplicity of the deep kernel machine loss to develop a new family of optimizers, based on a matrix equation from control theory, that converges in around 10 steps.

It is an open question whether preferences for visual art can be lawfully predicted from the basic constituent elements of a visual image. Here, we developed and tested a computational framework to investigate how aesthetic values are formed. We show that it is possible to explain human preferences for a visual art piece based on a mixture of low- and high-level features of the image. Subjective value ratings could be predicted not only within but also across individuals, using a regression model with a common set of interpretable features. We also show that the features predicting aesthetic preference can emerge hierarchically within a deep convolutional neural network trained only for object recognition. Our findings suggest that human preferences for art can be explained at least in part as a systematic integration over the underlying visual features of an image.

Artificial intelligence (AI) and machine learning (ML) can facilitate new paradigms and solutions in almost every research field. Collaboration is essential to achieve tangible and concrete progress in impactful and meaningful AI and ML research, due to its transdisciplinary nature. Come and meet University of Pretoria (UP) academics that are embracing and exploring the opportunities that AI and ML offer to transcend the conventional boundaries of their disciplines. Join the discussion to debate this new frontier of opportunities and challenges that may enable you to look beyond the obvious, and discover new directions and opportunities that we may offer for tomorrow — together!

What does an executive at a South African Bank, a machine learning lead, and a CEO of an AI company have in common? They all will be on a panel talking about careers in Data Science, Machine Learning and Artificial Intelligence

Deep learning has made tremendous progress in the last year but it's high computational and memory requirements impose challenges in using deep learning on edge devices. There has been some progress in lowering memory requirements of deep neural networks (for instance, use of half-precision) but there has been minimal effort in developing alternative efficient computational paradigms. Inspired by the brain, Spiking Neural Networks (SNN) provide an energy-efficient alternative to conventional rate-based neural networks. However, SNN architectures that employ the traditional feedforward and feedback pass do not fully exploit the asynchronous event-based processing paradigm of SNNs. In the first part of my talk, I will present my work on predictive coding which offers a fundamentally different approach to developing neural networks that are particularly suitable for event-based processing. In the second part of my talk, I will present our work on development of approaches for SNNs that target specific problems like low response latency and continual learning. References Dora, S., Bohte, S. M., & Pennartz, C. (2021). Deep Gated Hebbian Predictive Coding Accounts for Emergence of Complex Neural Response Properties Along the Visual Cortical Hierarchy. Frontiers in Computational Neuroscience, 65. Saranirad, V., McGinnity, T. M., Dora, S., & Coyle, D. (2021, July). DoB-SNN: A New Neuron Assembly-Inspired Spiking Neural Network for Pattern Classification. In 2021 International Joint Conference on Neural Networks (IJCNN) (pp. 1-6). IEEE. Machingal, P., Thousif, M., Dora, S., Sundaram, S., Meng, Q. (2021). A Cross Entropy Loss for Spiking Neural Networks. Expert Systems with Applications (under review).

We introduce Norse: An open-source library for gradient-based training of spiking neural networks. In contrast to neuron simulators which mainly target computational neuroscientists, our library seamlessly integrates with the existing PyTorch ecosystem using abstractions familiar to the machine learning community. This has immediate benefits in that it provides a familiar interface, hardware accelerator support and, most importantly, the ability to use gradient-based optimization. While many parallel efforts in this direction exist, Norse emphasizes flexibility and usability in three ways. Users can conveniently specify feed-forward (convolutional) architectures, as well as arbitrarily connected recurrent networks. We strictly adhere to a functional and class-based API such that neuron primitives and, for example, plasticity rules composes. Finally, the functional core API ensures compatibility with the PyTorch JIT and ONNX infrastructure. We have made progress to support network execution on the SpiNNaker platform and plan to support other neuromorphic architectures in the future. While the library is useful in its present state, it also has limitations we will address in ongoing work. In particular, we aim to implement event-based gradient computation, using the EventProp algorithm, which will allow us to support sparse event-based data efficiently, as well as work towards support of more complex neuron models. With this library, we hope to contribute to a joint future of computational neuroscience and neuromorphic computing.

Last year’s SNUFA workshop report concluded “Moving toward neuron numbers comparable with biology and applying these networks to real-world data-sets will require the development of novel algorithms, software libraries, and dedicated hardware accelerators that perform well with the specifics of spiking neural networks” [1]. Taking inspiration from machine learning libraries — where techniques such as parallel batch training minimise latency and maximise GPU occupancy — as well as our previous research on efficiently simulating SNNs on GPUs for computational neuroscience [2,3], we are extending our GeNN SNN simulator to pursue this vision. To explore GeNN’s potential, we use the eProp learning rule [4] — which approximates RTRL — to train SNN classifiers on the Spiking Heidelberg Digits and the Spiking Sequential MNIST datasets. We find that the performance of these classifiers is comparable to those trained using BPTT [5] and verify that the theoretical advantages of neuron models with adaptation dynamics [5] translate to improved classification performance. We then measured execution times and found that training an SNN classifier using GeNN and eProp becomes faster than SpyTorch and BPTT after less than 685 timesteps and much larger models can be trained on the same GPU when using GeNN. Furthermore, we demonstrate that our implementation of parallel batch training improves training performance by over 4⨉ and enables near-perfect scaling across multiple GPUs. Finally, we show that performing inference using a recurrent SNN using GeNN uses less energy and has lower latency than a comparable LSTM simulated with TensorFlow [6].

Norse aims to exploit the advantages of bio-inspired neural components, which are sparse and event-driven - a fundamental difference from artificial neural networks. Norse expands PyTorch with primitives for bio-inspired neural components, bringing you two advantages: a modern and proven infrastructure based on PyTorch and deep learning-compatible spiking neural network components.

Gradient-based optimization powered by the backpropagation algorithm proved to be the pivotal method in the training of non-spiking artificial neural networks. At the same time, spiking neural networks hold the promise for efficient processing of real-world sensory data by communicating using discrete events in continuous time. We derive the backpropagation algorithm for a recurrent network of spiking (leaky integrate-and-fire) neurons with hard thresholds and show that the backward dynamics amount to an event-based backpropagation of errors through time. Our derivation uses the jump conditions for partial derivatives at state discontinuities found by applying the implicit function theorem, allowing us to avoid approximations or substitutions. We find that the gradient exists and is finite almost everywhere in weight space, up to the null set where a membrane potential is precisely tangent to the threshold. Our presented algorithm, EventProp, computes the exact gradient with respect to a general loss function based on spike times and membrane potentials. Crucially, the algorithm allows for an event-based communication scheme in the backward phase, retaining the potential advantages of temporal sparsity afforded by spiking neural networks. We demonstrate the optimization of spiking networks using gradients computed via EventProp and the Yin-Yang and MNIST datasets with either a spike time-based or voltage-based loss function and report competitive performance. Our work supports the rigorous study of gradient-based optimization in spiking neural networks as well as the development of event-based neuromorphic architectures for the efficient training of spiking neural networks. While we consider the leaky integrate-and-fire model in this work, our methodology generalises to any neuron model defined as a hybrid dynamical system.

Recent advances in neuromorphic hardware and Surrogate Gradient (SG) learning highlight the potential of Spiking Neural Networks (SNNs) for energy-efficient signal processing and learning. Like in Artificial Neural Networks (ANNs), training performance in SNNs strongly depends on the initialization of synaptic and neuronal parameters. While there are established methods of initializing deep ANNs for high performance, effective strategies for optimal SNN initialization are lacking. Here, we address this gap and propose flexible data-dependent initialization strategies for SNNs.

The brain has a hugely diverse and heterogeneous nature. The exact role of heterogeneity has been relatively little explored as most neural models tend to be largely homogeneous. We trained spiking neural networks with varying degrees of heterogeneity on complex real-world tasks and found that heterogeneity resulted in more stable and robust training and improved training performance, especially for tasks with a higher temporal structure. Moreover, the optimal distribution of parameters found by training was found to be similar to experimental observations. These findings suggest that heterogeneity is not simply a result of noisy biological processes, but it may play a crucial role for learning in complex, changing environments.

Depth estimation is an important computer vision task, useful in particular for navigation in autonomous vehicles, or for object manipulation in robotics. Here we solved it using an end-to-end neuromorphic approach, combining two event-based cameras and a Spiking Neural Network (SNN) with a slightly modified U-Net-like encoder-decoder architecture, that we named StereoSpike. More specifically, we used the Multi Vehicle Stereo Event Camera Dataset (MVSEC). It provides a depth ground-truth, which was used to train StereoSpike in a supervised manner, using surrogate gradient descent. We propose a novel readout paradigm to obtain a dense analog prediction –the depth of each pixel– from the spikes of the decoder. We demonstrate that this architecture generalizes very well, even better than its non-spiking counterparts, leading to state-of-the-art test accuracy. To the best of our knowledge, it is the first time that such a large-scale regression problem is solved by a fully spiking network. Finally, we show that low firing rates (<10%) can be obtained via regularization, with a minimal cost in accuracy. This means that StereoSpike could be implemented efficiently on neuromorphic chips, opening the door for low power real time embedded systems.

A rich data representation that finds wide application in industry and research is the so-called knowledge graph - a graph-based structure where entities are depicted as nodes and relations between them as edges. Complex systems like molecules, social networks and industrial factory systems can be described using the common language of knowledge graphs, allowing the usage of graph embedding algorithms to make context-aware predictions in these information-packed environments.

This is my C-14 Impaler gauss rifle! There are many like it, but this one is mine!" - A terran marine If you have never heard of a terran marine before, then you have probably missed out on playing the very engaging and entertaining strategy computer game, StarCraft. However, don’t despair, because what we have in store might be even more exciting! In this interactive session, we will take you through, step-by-step, on how to train a team of terran marines to defeat a team of marines controlled by the built-in game AI in StarCraft II. How will we achieve this? Using multi-agent reinforcement learning (MARL). MARL is a useful framework for building distributed intelligent systems. In MARL, multiple agents are trained to act as individual decision-makers of some larger system, while learning to work as a team. We will show you how to use Mava (https://github.com/instadeepai/Mava), a newly released research framework for MARL to build a multi-agent learning system for StarCraft II. We will provide the necessary guidance, tools and background to understand the key concepts behind MARL, how to use Mava building blocks to build systems and how to train a system from scratch. We will conclude the session by briefly sharing various exciting real-world application areas for MARL at InstaDeep, such as large-scale autonomous train navigation and circuit board routing. These are problems that become exponentially more difficult to solve as they scale. Finally, we will argue that many of humanity’s most important practical problems are reminiscent of the ones just described. These include, for example, the need for sustainable management of distributed resources under the pressures of climate change, or efficient inventory control and supply routing in critical distribution networks, or robotic teams for rescue missions and exploration. We believe MARL has enormous potential to be applied in these areas and we hope to inspire you to get excited and interested in MARL and perhaps one day contribute to the field!

In the Learning Salon, we will discuss the similarities and differences between biological and machine learning, including individuals with diverse perspectives and backgrounds, so we can all learn from one another.

Large-scale neural recordings contain high-dimensional structure that cannot be easily captured by existing data visualization methods. We therefore developed an embedding algorithm called Rastermap, which captures highly nonlinear relationships between neurons, and provides useful visualizations by assigning each neuron to a location in the embedding space. Compared to standard algorithms such as t-SNE and UMAP, Rastermap finds finer and higher dimensional patterns of neural variability, as measured by quantitative benchmarks. We applied Rastermap to a variety of datasets, including spontaneous neural activity, neural activity during a virtual reality task, widefield neural imaging data during a 2AFC task, artificial neural activity from an agent playing atari games, and neural responses to visual textures. We found within these datasets unique subpopulations of neurons encoding abstract properties of the environment.

Tali's work emphasized the tradeoff between compression and information preservation. In this talk I will explore this theme in the context of deep learning. Artificial neural networks have recently revolutionized the field of machine learning. However, we still do not have sufficient theoretical understanding of how such models can be successfully learned. Two specific questions in this context are: how can neural nets be learned despite the non-convexity of the learning problem, and how can they generalize well despite often having more parameters than training data. I will describe our recent work showing that gradient-descent optimization indeed leads to 'simpler' models, where simplicity is captured by lower weight norm and in some cases clustering of weight vectors. We demonstrate this for several teacher and student architectures, including learning linear teachers with ReLU networks, learning boolean functions and learning convolutional pattern detection architectures.

Many tasks that are easy for humans are difficult for machines. In particular, while humans excel at tasks that require generalising across problems, machine systems notably struggle. One such task that has received a good amount of attention is the Synthetic Visual Reasoning Test (SVRT). The SVRT consists of a range of problems where simple visual stimuli must be categorised into one of two categories based on an unknown rule that must be induced. Conventional machine learning approaches perform well only when trained to categorise based on a single rule and are unable to generalise without extensive additional training to tasks with any additional rules. Multiple theories of higher-level cognition posit that humans solve such tasks using structured relational representations. Specifically, people learn rules based on structured representations that generalise to novel instances quickly and easily. We believe it is possible to model this approach in a single system which learns all the required relational representations from scratch and performs tasks such as SVRT in a single run. Here, we present a system which expands the DORA/LISA architecture and augments the existing model with principally novel components, namely a) visual reasoning based on the established theories of recognition by components; b) the process of learning complex relational representations by synthesis (in addition to learning by analysis). The proposed augmented model matches human behaviour on SVRT problems. Moreover, the proposed system stands as perhaps a more realistic account of human cognition, wherein rather than using tools that has been shown successful in the machine learning field to inform psychological theorising, we use established psychological theories to inform developing a machine system.