Ensembles of deep neural networks are known to achieve state-of-the-art performance in uncertainty estimation and lead to accuracy improvement. In this work, we focus on a classification problem and investigate the behavior of both non-calibrated and calibrated negative log-likelihood (CNLL) of a deep ensemble as a function of the ensemble size and the member network size. We indicate the conditions under which CNLL follows a power law w.r.t. ensemble size or member network size, and analyze the dynamics of the parameters of the discovered power laws. Our important practical finding is that one large network may perform worse than an ensemble of several medium-size networks with the same total number of parameters (we call this ensemble a memory split). Using the detected power law-like dependencies, we can predict (1) the possible gain from the ensembling of networks with given structure, (2) the optimal memory split given a memory budget, based on a relatively small number of trained networks. We describe the memory split advantage effect in more details in arXiv:2005.07292

Geometric structure of an optimization landscape is argued to be fundamentally important to support the success of deep learning. However, recent research efforts focused on either of toy random models with unrealistic assumptions and numerical evidences about different shapes of the optimization landscape, thereby lacking a unified view about the nature of the landscape. Here, we propose a statistical mechanics framework by directly building a least structured model of the high-dimensional weight space, considering realistic structured data, stochastic gradient descent algorithms, and the computational depth of the network parametrized by weight parameters. We also consider whether the number of network parameters outnumbers the number of supplied training data, namely, over- or under-parametrization. Our least structured model predicts that the weight spaces of the under-parametrization and over-parameterization cases belong to the same class. These weight spaces are well-connected without any heterogeneous geometric properties. In contrast, the shallow-network has a shattered weight space, characterized by discontinuous phase transitions in physics, thereby clarifying roles of depth in deep learning. Our effective model also predicts that inside a deep network, there exists a liquid-like central part of the architecture in the sense that the weights in this part behave as randomly as possible. Our work may thus explain why deep learning is unreasonably effective in terms of the high-dimensional weight space, and how deep networks are different from shallow ones.

In the field of non-monotonic logics, the notion of Rational Closure (RC) is acknowledged as a prominent approach. In recent years, RC has gained even more popularity in the context of Description Logics (DLs), the logic underpinning the semantic web standard ontology language OWL 2, whose main ingredients are classes and roles. In this work, we show how to integrate RC within the triple language RDFS, which together with OWL2 are the two major standard semantic web ontology languages. To do so, we start from $\rho df$, which is the logic behind RDFS, and then extend it to $\rho df_\bot$, allowing to state that two entities are incompatible. Eventually, we propose defeasible $\rho df_\bot$ via a typical RC construction. The main features of our approach are: (i) unlike most other approaches that add an extra non-monotone rule layer on top of monotone RDFS, defeasible $\rho df_\bot$ remains syntactically a triple language and is a simple extension of $\rho df_\bot$ by introducing some new predicate symbols with specific semantics. In particular, any RDFS reasoner/store may handle them as ordinary terms if it does not want to take account for the extra semantics of the new predicate symbols; (ii) the defeasible $\rho df_\bot$ entailment decision procedure is build on top of the $\rho df_\bot$ entailment decision procedure, which in turn is an extension of the one for $\rho df$ via some additional inference rules favouring an potential implementation; and (iii) defeasible $\rho df_\bot$ entailment can be decided in polynomial time.

We propose Graph Contrastive Learning (GraphCL), a general framework for learning node representations in a self supervised manner. GraphCL learns node embeddings by maximizing the similarity between the representations of two randomly perturbed versions of the intrinsic features and link structure of the same node's local subgraph. We use graph neural networks to produce two representations of the same node and leverage a contrastive learning loss to maximize agreement between them. In both transductive and inductive learning setups, we demonstrate that our approach significantly outperforms the state-of-the-art in unsupervised learning on a number of node classification benchmarks.

Anti-money laundering (AML) is expensive and labour-intensive, and artificial intelligence (AI) can offer improved efficiency gains. Could they be a match made in heaven? This Risk.net webinar, in association with NICE Actimize, took place amid the strain on banks' back offices driven by the lockdown in response to the global Covid‑19 pandemic, and explores this potential pairing Today's evolving regulatory environment and criminal typologies have influenced AML compliance teams to adopt AI technologies such as machine learning to improve detection and better focus analyst workloads. The marriage of AI to existing compliance processes and risk modelling techniques has the potential to eliminate backlogs and create new efficiencies. But there may be some risks and question marks for those in the early stages of adoption. The strain on many financial institutions has only increased in 2020 due to the unexpected arrival of Covid‑19.

Scientists have recorded the first documented evidence of a great white shark attacking and killing an enormous humpback whale. Video taken from a drone off the coast of South Africa shows the 13-foot long shark hunting the whale which was around 33ft long and in ill health. Ryan Johnson, a marine biologist who observed the massacre, says the ordeal lasted about 50 minutes before the whale eventually died. Mr Johnson says the great white was very strategic in taking down the behemoth, and targeted its most vulnerable area, on the tail, before drowning the ailing whale. The great white is believed to be a shark called Helen which was named and tagged as part of a 2013 study by Mr Johnson.

The Machine Learning report provides independent information about the Machine Learning industry supported by extensive research on factors such as industry segments size & trends, inhibitors, dynamics, drivers, opportunities & challenges, environment & policy, cost overview, porter's five force analysis, and key companies profiles including business overview and recent development. The research report on Machine Learning market thoroughly investigates historical data of this business sphere to lay out the future roadmap of the industry. The study attempts to predict a long-term picture of the market scenario with respect to the various growth indicators, hindrances, and opportunities that determine the industry expansion. Moreover, the report provides an exhaustive synopsis of the industry at a global and regional level. In addition, it covers the impact of COVID-19 pandemic on the leading industry players and various market segmentations.

Despite the success of existing tensor factorization methods, most of them conduct a multilinear decomposition, and rarely exploit powerful modeling frameworks, like deep neural networks, to capture a variety of complicated interactions in data. More important, for highly expressive, deep factorization, we lack an effective approach to handle streaming data, which are ubiquitous in real-world applications. To address these issues, we propose SPIDER, a Streaming ProbabilistIc Deep tEnsoR factorization method. We first use Bayesian neural networks (NNs) to construct a deep tensor factorization model. We assign a spike-and-slab prior over the NN weights to encourage sparsity and prevent overfitting. We then use Taylor expansions and moment matching to approximate the posterior of the NN output and calculate the running model evidence, based on which we develop an efficient streaming posterior inference algorithm in the assumed-density-filtering and expectation propagation framework. Our algorithm provides responsive incremental updates for the posterior of the latent factors and NN weights upon receiving new tensor entries, and meanwhile select and inhibit redundant/useless weights. We show the advantages of our approach in four real-world applications.

Characterising intractable high-dimensional random variables is one of the fundamental challenges in stochastic computation. The recent surge of transport maps offers a mathematical foundation and new insights for tackling this challenge by coupling intractable random variables with tractable reference random variables. This paper generalises a recently developed functional tensor-train (FTT) approximation of the inverse Rosenblatt transport [14] to a wide class of high-dimensional nonnegative functions, such as unnormalised probability density functions. First, we extend the inverse Rosenblatt transform to enable the transport to general reference measures other than the uniform measure. We develop an efficient procedure to compute this transport from a squared FTT decomposition which preserves the monotonicity. More crucially, we integrate the proposed monotonicity-preserving FTT transport into a nested variable transformation framework inspired by deep neural networks. The resulting deep inverse Rosenblatt transport significantly expands the capability of tensor approximations and transport maps to random variables with complicated nonlinear interactions and concentrated density functions. We demonstrate the efficacy of the proposed approach on a range of applications in statistical learning and uncertainty quantification, including parameter estimation for dynamical systems and inverse problems constrained by partial differential equations.

In this paper, we investigate the limiting behavior of a continuous-time counterpart of the Stochastic Gradient Descent (SGD) algorithm applied to two-layer overparameterized neural networks, as the number or neurons (ie, the size of the hidden layer) $N \to +\infty$. Following a probabilistic approach, we show 'propagation of chaos' for the particle system defined by this continuous-time dynamics under different scenarios, indicating that the statistical interaction between the particles asymptotically vanishes. In particular, we establish quantitative convergence with respect to $N$ of any particle to a solution of a mean-field McKean-Vlasov equation in the metric space endowed with the Wasserstein distance. In comparison to previous works on the subject, we consider settings in which the sequence of stepsizes in SGD can potentially depend on the number of neurons and the iterations. We then identify two regimes under which different mean-field limits are obtained, one of them corresponding to an implicitly regularized version of the minimization problem at hand. We perform various experiments on real datasets to validate our theoretical results, assessing the existence of these two regimes on classification problems and illustrating our convergence results.