Most state-of-the-art Graph Neural Networks (GNNs) can be defined as a form of graph convolution which can be realized by message passing between direct neighbors or beyond. To scale such GNNs to large graphs, various neighbor-, layer-, or subgraph-sampling techniques are proposed to alleviate the "neighbor explosion" problem by considering only a small subset of messages passed to the nodes in a mini-batch. However, sampling-based methods are difficult to apply to GNNs that utilize many-hops-away or global context each layer, show unstable performance for different tasks and datasets, and do not speed up model inference. We propose a principled and fundamentally different approach, VQ-GNN, a universal framework to scale up any convolution-based GNNs using Vector Quantization (VQ) without compromising the performance. In contrast to sampling-based techniques, our approach can effectively preserve all the messages passed to a mini-batch of nodes by learning and updating a small number of quantized reference vectors of global node representations, using VQ within each GNN layer. Our framework avoids the "neighbor explosion" problem of GNNs using quantized representations combined with a low-rank version of the graph convolution matrix. We show that such a compact low-rank version of the gigantic convolution matrix is sufficient both theoretically and experimentally. In company with VQ, we design a novel approximated message passing algorithm and a nontrivial back-propagation rule for our framework. Experiments on various types of GNN backbones demonstrate the scalability and competitive performance of our framework on large-graph node classification and link prediction benchmarks.

The nature of available economic data has changed fundamentally in the last decade due to the economy's digitisation. With the prevalence of often black box data-driven machine learning methods, there is a necessity to develop interpretable machine learning methods that can conduct econometric inference, helping policymakers leverage the new nature of economic data. We therefore present a novel Variational Bayesian Inference approach to incorporate a time-varying parameter auto-regressive model which is scalable for big data. Our model is applied to a large blockchain dataset containing prices, transactions of individual actors, analyzing transactional flows and price movements on a very granular level. The model is extendable to any dataset which can be modelled as a dynamical system. We further improve the simple state-space modelling by introducing non-linearities in the forward model with the help of machine learning architectures.

Traffic forecasting is an integral part of intelligent transportation systems (ITS). Achieving a high prediction accuracy is a challenging task due to a high level of dynamics and complex spatial-temporal dependency of road networks. For this task, we propose Graph Attention-Convolution-Attention Networks (GACAN). The model uses a novel Att-Conv-Att (ACA) block which contains two graph attention layers and one spectral-based GCN layer sandwiched in between. The graph attention layers are meant to capture temporal features while the spectral-based GCN layer is meant to capture spatial features. The main novelty of the model is the integration of time series of four different time granularities: the original time series, together with hourly, daily, and weekly time series. Unlike previous work that used multi-granularity time series by handling every time series separately, GACAN combines the outcome of processing all time series after each graph attention layer. Thus, the effects of different time granularities are integrated throughout the model. We perform a series of experiments on three real-world datasets. The experimental results verify the advantage of using multi-granularity time series and that the proposed GACAN model outperforms the state-of-the-art baselines.

Knowledge Distillation (KD) utilizes training data as a transfer set to transfer knowledge from a complex network (Teacher) to a smaller network (Student). Several works have recently identified many scenarios where the training data may not be available due to data privacy or sensitivity concerns and have proposed solutions under this restrictive constraint for the classification task. Unlike existing works, we, for the first time, solve a much more challenging problem, i.e., "KD for object detection with zero knowledge about the training data and its statistics". Our proposed approach prepares pseudo-targets and synthesizes corresponding samples (termed as "Multi-Object Impressions"), using only the pretrained Faster RCNN Teacher network. We use this pseudo-dataset as a transfer set to conduct zero-shot KD for object detection. We demonstrate the efficacy of our proposed method through several ablations and extensive experiments on benchmark datasets like KITTI, Pascal and COCO. Our approach with no training samples, achieves a respectable mAP of 64.2% and 55.5% on the student with same and half capacity while performing distillation from a Resnet-18 Teacher of 73.3% mAP on KITTI.

Many applications of generative models rely on the marginalization of their high-dimensional output probability distributions. Normalization functions that yield sparse probability distributions can make exact marginalization more computationally tractable. However, sparse normalization functions usually require alternative loss functions for training since the log-likelihood is undefined for sparse probability distributions. Furthermore, many sparse normalization functions often collapse the multimodality of distributions. In this work, we present $\textit{ev-softmax}$, a sparse normalization function that preserves the multimodality of probability distributions. We derive its properties, including its gradient in closed-form, and introduce a continuous family of approximations to $\textit{ev-softmax}$ that have full support and can be trained with probabilistic loss functions such as negative log-likelihood and Kullback-Leibler divergence. We evaluate our method on a variety of generative models, including variational autoencoders and auto-regressive architectures. Our method outperforms existing dense and sparse normalization techniques in distributional accuracy. We demonstrate that $\textit{ev-softmax}$ successfully reduces the dimensionality of probability distributions while maintaining multimodality.

This paper studies how the dataset may be the cause of the anomalous generalization performance of deep networks. We show that the data correlation matrix of typical classification datasets has an eigenspectrum where, after a sharp initial drop, a large number of small eigenvalues are distributed uniformly over an exponentially large range. This structure is mirrored in a network trained on this data: we show that the Hessian and the Fisher Information Matrix (FIM) have eigenvalues that are spread uniformly over exponentially large ranges. We call such eigenspectra "sloppy" because sets of weights corresponding to small eigenvalues can be changed by large magnitudes without affecting the loss. Networks trained on atypical, non-sloppy synthetic data do not share these traits. We show how this structure in the data can give to non-vacuous PAC-Bayes generalization bounds analytically; we also construct data-distribution dependent priors that lead to accurate bounds using numerical optimization.

Self-supervised learning has been shown to be very effective in learning useful representations, and yet much of the success is achieved in data types such as images, audio, and text. The success is mainly enabled by taking advantage of spatial, temporal, or semantic structure in the data through augmentation. However, such structure may not exist in tabular datasets commonly used in fields such as healthcare, making it difficult to design an effective augmentation method, and hindering a similar progress in tabular data setting. In this paper, we introduce a new framework, Subsetting features of Tabular data (SubTab), that turns the task of learning from tabular data into a multi-view representation learning problem by dividing the input features to multiple subsets. We argue that reconstructing the data from the subset of its features rather than its corrupted version in an autoencoder setting can better capture its underlying latent representation. In this framework, the joint representation can be expressed as the aggregate of latent variables of the subsets at test time, which we refer to as collaborative inference. Our experiments show that the SubTab achieves the state of the art (SOTA) performance of 98.31% on MNIST in tabular setting, on par with CNN-based SOTA models, and surpasses existing baselines on three other real-world datasets by a significant margin.

Copulas are a powerful tool for modeling multivariate distributions as they allow to separately estimate the univariate marginal distributions and the joint dependency structure. However, known parametric copulas offer limited flexibility especially in high dimensions, while commonly used non-parametric methods suffer from the curse of dimensionality. A popular remedy is to construct a tree-based hierarchy of conditional bivariate copulas. In this paper, we propose a flexible, yet conceptually simple alternative based on implicit generative neural networks. The key challenge is to ensure marginal uniformity of the estimated copula distribution. We achieve this by learning a multivariate latent distribution with unspecified marginals but the desired dependency structure. By applying the probability integral transform, we can then obtain samples from the high-dimensional copula distribution without relying on parametric assumptions or the need to find a suitable tree structure. Experiments on synthetic and real data from finance, physics, and image generation demonstrate the performance of this approach.

Why you should read it: Widely considered a classic, this book is a comprehensive look at AI as a field and technology. The book is now on its fourth edition, and covers machine learning, deep learning, and other types of AI, as well as important concerns around AI safety and privacy.

ML needs to account for uncertainty! Have you heard of probabilistic programming? It's a programming paradigm in which you can easily specify probabilistic models and perform inference on them. These languages greatly simplify the task of creating systems that handle uncertainty. For example, Pyro (from Uber AI Labs) enables flexible and expressive deep probabilistic modeling, unifying the best of modern deep learning and Bayesian modeling. And while we won't touch on probabilistic programming in this tutorial, you may want to know why probabilistic approaches are so needed in ML and why these languages are growing so quickly.