arXiv research

Recent work in signal propagation theory has shown that dropout limits the depth to which information can propagate through a neural network. In this paper, we investigate the effect of initialisation on training speed and generalisation for ReLU networks within this depth limit. We ask the following research question:…

Optimistic initialisation is an effective strategy for efficient exploration in reinforcement learning (RL). In the tabular case, all provably efficient model-free algorithms rely on it. However, model-free deep RL algorithms do not use optimistic initialisation despite taking inspiration from these provably efficient …

New weight initialisation for ICNNs accelerates learning and improves generalization.

The training of stochastic neural network models with binary ($\pm1$) weights and activations via continuous surrogate networks is investigated. We derive new surrogates using a novel derivation based on writing the stochastic neural network as a Markov chain. This derivation also encompasses existing variants of the s…

Unified theory explains two failure modes of deep transformers and provides initialisation guidelines.

Stochastic regularisation is an important weapon in the arsenal of a deep learning practitioner. However, despite recent theoretical advances, our understanding of how noise influences signal propagation in deep neural networks remains limited. By extending recent work based on mean field theory, we develop a new frame…

This paper presents a new way of selecting an initial solution for the k-modes algorithm that allows for a notion of mathematical fairness and a leverage of the data that the common initialisations from literature do not. The method, which utilises the Hospital-Resident Assignment Problem to find the set of initial clu…

K-Means is one of the most used algorithms for data clustering and the usual clustering method for benchmarking. Despite its wide application it is well-known that it suffers from a series of disadvantages; it is only able to find local minima and the positions of the initial clustering centres (centroids) can greatly …

We present a general method for fitting finite mixture models (FMM). Learning in a mixture model consists of finding the most likely cluster assignment for each data-point, as well as finding the parameters of the clusters themselves. In many mixture models, this is difficult with current learning methods, where the mo…

Early alignment in neural networks leads to sparse representations but hinders convergence.

In this paper we make two novel contributions to hierarchical clustering. First, we introduce an anomalous pattern initialisation method for hierarchical clustering algorithms, called A-Ward, capable of substantially reducing the time they take to converge. This method generates an initial partition with a sufficiently…

Analyzes deep neural networks training errors with SGD and random init.

Recent work has established the equivalence between deep neural networks and Gaussian processes (GPs), resulting in so-called neural network Gaussian processes (NNGPs). The behaviour of these models depends on the initialisation of the corresponding network. In this work, we consider the impact of noise regularisation …

This study explains gradient flow dynamics in neural networks for small initialisation.

Two algorithms converge to dictionary learning with geometric rate for non-uniform data.

We present a probabilistic framework for both (i) determining the initial settings of kernel adaptive filters (KAFs) and (ii) constructing fully-adaptive KAFs whereby in addition to weights and dictionaries, kernel parameters are learnt sequentially. This is achieved by formulating the estimator as a probabilistic mode…

A new method averages neural network parameters to rank features robustly.

Deep reinforcement learning agents have recently been successful across a variety of discrete and continuous control tasks; however, they can be slow to train and require a large number of interactions with the environment to learn a suitable policy. This is borne out by the fact that a reinforcement learning agent has…

New approach connects quantum phases to VQA trainability, enabling better scaling.

The aim of this paper is to provide a general mathematical framework for group equivariance in the machine learning context. The framework builds on a synergy between persistent homology and the theory of group actions. We define group-equivariant non-expansive operators (GENEOs), which are maps between function spaces…

The paper extends infinite-width analysis to neural network Jacobians, revealing convergence to Gaussian processes and linear ODEs.

GENESIS-V2 infers unordered object representations without iterative refinement.

In this paper we propose a Bayesian method for estimating architectural parameters of neural networks, namely layer size and network depth. We do this by learning concrete distributions over these parameters. Our results show that regular networks with a learnt structure can generalise better on small datasets, while f…

einspace expands NAS search space to include diverse neural architectures.

In this paper we propose a model that combines the strengths of RNNs and SGVB: the Variational Recurrent Auto-Encoder (VRAE). Such a model can be used for efficient, large scale unsupervised learning on time series data, mapping the time series data to a latent vector representation. The model is generative, such that …

Graph classification receives a great deal of attention from the non-Euclidean machine learning community. Recent advances in graph coarsening have enabled the training of deeper networks and produced new state-of-the-art results in many benchmark tasks. We examine how these architectures train and find that performanc…

We propose the application of a semi-supervised learning method to improve the performance of acoustic modelling for automatic speech recognition based on deep neural net- works. As opposed to unsupervised initialisation followed by supervised fine tuning, our method takes advantage of both unlabelled and labelled data…

The choice of sentence encoder architecture reflects assumptions about how a sentence's meaning is composed from its constituent words. We examine the contribution of these architectures by holding them randomly initialised and fixed, effectively treating them as as hand-crafted language priors, and evaluating the resu…

Study combines variational inference and transformers for seasonal climate predictions.

Transformers interpreted as probabilistic Laplacian Eigenmaps steps.

Fine-tuning neural networks is widely used to transfer valuable knowledge from high-resource to low-resource domains. In a standard fine-tuning scheme, source and target problems are trained using the same architecture. Although capable of adapting to new domains, pre-trained units struggle with learning uncommon targe…

Heuristic optimisers which search for an optimal configuration of variables relative to an objective function often get stuck in local optima where the algorithm is unable to find further improvement. The standard approach to circumvent this problem involves periodically restarting the algorithm from random initial con…

We show that the stochasticity in training ResNets for image classification on GPUs in TensorFlow is dominated by the non-determinism from GPUs, rather than by the initialisation of the weights and biases of the network or by the sequence of minibatches given. The standard deviation of test set accuracy is 0.02 with fi…

We demonstrate the first application of deep reinforcement learning to autonomous driving. From randomly initialised parameters, our model is able to learn a policy for lane following in a handful of training episodes using a single monocular image as input. We provide a general and easy to obtain reward: the distance …

Activation functions play a key role in neural networks so it becomes fundamental to understand their advantages and disadvantages in order to achieve better performances. This paper will first introduce common types of non linear activation functions that are alternative to the well known sigmoid function and then eva…

Hierarchical pretraining with slow-fast ODEs

A new heuristic LM algorithm improves $k$-segmentation accuracy with less computation.

Gaussian processes are typically used for smoothing and interpolation on small datasets. We introduce a new Bayesian nonparametric framework -- GPatt -- enabling automatic pattern extrapolation with Gaussian processes on large multidimensional datasets. GPatt unifies and extends highly expressive kernels and fast exact…

Wide stochastic networks show Gaussian behavior and improve training with PAC-Bayesian methods.

Bayesian inference for factorial hidden Markov models is challenging due to the exponentially sized latent variable space. Standard Monte Carlo samplers can have difficulties effectively exploring the posterior landscape and are often restricted to exploration around localised regions that depend on initialisation. We …

Optimizes wide low-rank neural networks for reduced parameters and cost.

High-dimensional unimodal distributions can cause MCMC methods to fail.

We provide an overview of several non-linear activation functions in a neural network architecture that have proven successful in many machine learning applications. We conduct an empirical analysis on the effectiveness of using these function on the MNIST classification task, with the aim of clarifying which functions…

We propose a novel approximate inference algorithm that approximates a target distribution by amortising the dynamics of a user-selected MCMC sampler. The idea is to initialise MCMC using samples from an approximation network, apply the MCMC operator to improve these samples, and finally use the samples to update the a…

Paper uses variational inference to estimate nonlinear models.

We introduce a new method for sparse principal component analysis, based on the aggregation of eigenvector information from carefully-selected axis-aligned random projections of the sample covariance matrix. Unlike most alternative approaches, our algorithm is non-iterative, so is not vulnerable to a bad choice of init…

Ensembles of neural networks (NNs) have long been used to estimate predictive uncertainty; a small number of NNs are trained from different initialisations and sometimes on differing versions of the dataset. The variance of the ensemble's predictions is interpreted as its epistemic uncertainty. The appeal of ensembling…

Study on neural networks with regularisation and its impact on training dynamics.