She's a robot who uses artificial intelligence to write and share hair-raising horror stories. Named after famed "Frankenstein" writer Mary Shelley, the bot is the brainchild of MIT postdoctoral candidate Pinar Yanardag and research scientist Manuel Cebrian. They claim that Shelley is the world's first collaborative AI horror writer. On Oct. 27, Shelley came to life. But, before she did, co-creators Pinar Yanardag and Manuel Cebrian spent a lot of time figuring out where to obtain enough data to power Shelley.

Come see Anshuman Guha, Data Scientist from Spark Cognition Speak at ODSC West. In machine learning, the most traditional and popular methods of clustering are hierarchical clustering (similarity-based clustering) and k-means clustering (feature-based clustering). Hierarchical clustering, put simply, is grouping together points in a vector space that are closest in distance from each other. Hierarchical clustering works great on small datasets. A major advantage of this method is the user does not need to know anything about the dataset in advance and specify any hyper-parameters (like number of clusters).

Was this the week that the space age vision of a car that drives itself became a reality? And are claims that artificial intelligence can transform healthcare a bit overhyped? On this week's Tech Tent podcast we explore the potential and limits of technology in health and transport. This week we woke up to the fact that autonomous cars could be with us sooner than we thought. That was the message from John Krafcik, chief executive of Waymo, the self-driving car division of Google - or Alphabet as we must learn to call it.

Future Decoded is a conference for customers and partners of the tech giant which was held in London over two days at the start of this month. Discussions about the potential of artificial intelligence (AI) were much in evidence. Microsoft's approach to AI is as a service delivered from its Azure cloud delivered through a network of large scale data centres and smaller edge based technologies. Also on the agenda was Microsoft's Quantum Computing software programme development which uses what is known as Topological Qubits code, described as "a robust type of quantum bit that Microsoft believes will serve as the basis for a scalable, general purpose quantum computer system." Microsoft plans to release Quantum Computing developer code before then end of the year.

The worldwide surge of multiresistant microbial strains has propelled the search for alternative treatment options. The study of Protein-Protein Interactions (PPIs) has been a cornerstone in the clarification of complex physiological and pathogenic processes, thus being a priority for the identification of vital components and mechanisms in pathogens. Despite the advances of laboratorial techniques, computational models allow the screening of protein interactions between entire proteomes in a fast and inexpensive manner. Here, we present a supervised machine learning model for the prediction of PPIs based on the protein sequence. We cluster amino acids regarding their physicochemical properties, and use the discrete cosine transform to represent protein sequences. A mesh of classifiers was constructed to create hyper-specialised classifiers dedicated to the most relevant pairs of molecular function annotations from Gene Ontology. Based on an exhaustive evaluation that includes datasets with different configurations, cross-validation and out-of-sampling validation, the obtained results outscore the state-of-the-art for sequence-based methods. For the final mesh model using SVM with RBF, a consistent average AUC of 0.84 was attained.

When used as a surrogate objective for maximum likelihood estimation in latent variable models, the evidence lower bound (ELBO) produces state-of-the-art results. Inspired by this, we consider the extension of the ELBO to a family of lower bounds defined by a particle filter's estimator of the marginal likelihood, the filtering variational objectives (FIVOs). FIVOs take the same arguments as the ELBO, but can exploit a model's sequential structure to form tighter bounds. We present results that relate the tightness of FIVO's bound to the variance of the particle filter's estimator by considering the generic case of bounds defined as log-transformed likelihood estimators. Experimentally, we show that training with FIVO results in substantial improvements over training the same model architecture with the ELBO on sequential data.

Epileptic seizure activity shows complicated dynamics in both space and time. To understand the evolution and propagation of seizures spatially extended sets of data need to be analysed. We have previously described an efficient filtering scheme using variational Laplace that can be used in the Dynamic Causal Modelling (DCM) framework [Friston, 2003] to estimate the temporal dynamics of seizures recorded using either invasive or non-invasive electrical recordings (EEG/ECoG). Spatiotemporal dynamics are modelled using a partial differential equation -- in contrast to the ordinary differential equation used in our previous work on temporal estimation of seizure dynamics [Cooray, 2016]. We provide the requisite theoretical background for the method and test the ensuing scheme on simulated seizure activity data and empirical invasive ECoG data. The method provides a framework to assimilate the spatial and temporal dynamics of seizure activity, an aspect of great physiological and clinical importance.

Sparse pseudo-point approximations for Gaussian process (GP) models provide a suite of methods that support deployment of GPs in the large data regime and enable analytic intractabilities to be sidestepped. However, the field lacks a principled method to handle streaming data in which both the posterior distribution over function values and the hyperparameter estimates are updated in an online fashion. The small number of existing approaches either use suboptimal hand-crafted heuristics for hyperparameter learning, or suffer from catastrophic forgetting or slow updating when new data arrive. This paper develops a new principled framework for deploying Gaussian process probabilistic models in the streaming setting, providing methods for learning hyperparameters and optimising pseudo-input locations. The proposed framework is assessed using synthetic and real-world datasets.

Learning an encoding of feature vectors in terms of an over-complete dictionary or a information geometric (Fisher vectors) construct is wide-spread in statistical signal processing and computer vision. In content based information retrieval using deep-learning classifiers, such encodings are learnt on the flattened last layer, without adherence to the multi-linear structure of the underlying feature tensor. We illustrate a variety of feature encodings incl. sparse dictionary coding and Fisher vectors along with proposing that a structured tensor factorization scheme enables us to perform retrieval that can be at par, in terms of average precision, with Fisher vector encoded image signatures. In short, we illustrate how structural constraints increase retrieval fidelity.

Motivated by applications in declarative data analysis, we study $\mathit{Datalog}_{\mathbb{Z}}$---an extension of positive Datalog with arithmetic functions over integers. This language is known to be undecidable, so we propose two fragments. In $\mathit{limit}~\mathit{Datalog}_{\mathbb{Z}}$ predicates are axiomatised to keep minimal/maximal numeric values, allowing us to show that fact entailment is coNExpTime-complete in combined, and coNP-complete in data complexity. Moreover, an additional $\mathit{stability}$ requirement causes the complexity to drop to ExpTime and PTime, respectively. Finally, we show that stable $\mathit{Datalog}_{\mathbb{Z}}$ can express many useful data analysis tasks, and so our results provide a sound foundation for the development of advanced information systems.