Sheffield ML

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.

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).

The success of deep learning sparked interest in whether the brain learns by using similar techniques for assigning credit to each synaptic weight for its contribution to the network output. However, the majority of current attempts at biologically-plausible learning methods are either non-local in time, require highly specific connectivity motives, or have no clear link to any known mathematical optimization method. Here, we introduce Deep Feedback Control (DFC), a new learning method that uses a feedback controller to drive a deep neural network to match a desired output target and whose control signal can be used for credit assignment. The resulting learning rule is fully local in space and time and approximates Gauss-Newton optimization for a wide range of feedback connectivity patterns. To further underline its biological plausibility, we relate DFC to a multi-compartment model of cortical pyramidal neurons with a local voltage-dependent synaptic plasticity rule, consistent with recent theories of dendritic processing. By combining dynamical system theory with mathematical optimization theory, we provide a strong theoretical foundation for DFC that we corroborate with detailed results on toy experiments and standard computer-vision benchmarks.

Climate change is one of our generation's greatest challenges, with increasingly severe consequences on global ecosystems and populations. Machine Learning has the potential to address many important challenges in climate change, from both mitigation (reducing its extent) and adaptation (preparing for unavoidable consequences) aspects. To present the extent of these opportunities, I will describe some of the projects that I am involved in, spanning from generative model to computer vision and natural language processing. There are many opportunities for fundamental innovation in this field, advancing the state-of-the-art in Machine Learning while ensuring that this fundamental progress translates into positive real-world impact.

While meta-learning algorithms are often viewed as algorithms that learn to learn, an alternative viewpoint frames meta-learning as inferring a hidden task variable from experience consisting of observations and rewards. From this perspective, learning to learn is learning to infer. This viewpoint can be useful in solving problems in meta-RL, which I’ll demonstrate through two examples: (1) enabling off-policy meta-learning, and (2) performing efficient meta-RL from image observations. I’ll also discuss how this perspective leads to an algorithm for few-shot image segmentation.

Poor air quality in cities is a significant threat to health and life expectancy, with over 80% of people living in urban areas exposed to air quality levels that exceed World Health Organisation limits. In this session, I present a multi-resolution multi-task framework that handles evidence integration under varying spatio-temporal sampling resolution and noise levels. We have developed both shallow Gaussian Process (GP) mixture models and deep GP constructions that naturally handle this evidence integration, as well as biases in the mean. These models underpin our work at the Alan Turing Institute towards providing spatio-temporal forecasts of air pollution across London. We demonstrate the effectiveness of our framework on both synthetic examples and applications on London air quality. For further information go to: https://www.turing.ac.uk/research/research-projects/london-air-quality. Collaborators: Oliver Hamelijnck, Theodoros Damoulas, Kangrui Wang and Mark Girolami.

Despite the consistency of symptoms at the cognitive level, we now know that brain disorders like Autism and Schizophrenia can each arise from mutations in >100 different genes. Presumably there is a convergence of “symptoms” at the level of neural circuits in diagnosed individuals. In this talk I will argue that redundancy in neural circuit parameters implies that we should take a circuit-function rather that circuit-component approach to understanding these disorders. Then I will present our recent empirical work testing a circuit-function theory for Autism: the idea that neural circuits show excess trial-to-trial variability in response to sensory stimuli, and instability in the representations across a timescale of days. For this we analysed in vivo neural population activity data recorded from somatosensory cortex of mouse models of Fragile-X syndrome, a disorder related to autism. Work with Beatriz Mizusaki (Univ of Bristol), Nazim Kourdougli, Anand Suresh, and Carlos Portera-Cailliau (Univ of California, Los Angeles).

The curse of dimensionality plagues models of reinforcement learning and decision-making. The process of abstraction solves this by constructing abstract variables describing features shared by different specific instances, reducing dimensionality and enabling generalization in novel situations. We characterized neural representations in monkeys performing a task where a hidden variable described the temporal statistics of stimulus-response-outcome mappings. Abstraction was defined operationally using the generalization performance of neural decoders across task conditions not used for training. This type of generalization requires a particular geometric format of neural representations. Neural ensembles in dorsolateral pre-frontal cortex, anterior cingulate cortex and hippocampus, and in simulated neural networks, simultaneously represented multiple hidden and explicit variables in a format reflecting abstraction. Task events engaging cognitive operations modulated this format. These findings elucidate how the brain and artificial systems represent abstract variables, variables critical for generalization that in turn confers cognitive flexibility.

Complex systems can be modeled at various levels of granularity, e.g., we can model a person at the cognitive level, on the neuronal level, or down to the biochemical level. When multiple models represent the same system at different scales, we would like to be able to reason about the causal effects of interventions on each level in such a way that the models remain consistent across levels. In the first part of this talk, I consider which conditions must be fulfilled for two structural equation models (SEMs) to stand in such a causally consistent relation. In the second part of the talk, I present recent work on learning causally consistent SEMs across multiple levels, distinguishing between bottom-up (micro- to macro-level) and top-down (macro- to micro-level) approaches.