Despite ample evidence that our concepts, our cognitive architecture, and mathematics itself are all deeply compositional, few models take advantage of this structure. We therefore propose a radically compositional approach to computational neuroscience, drawing on the methods of applied category theory. We describe how these tools grant us a means to overcome complexity and improve interpretability, and supply a rigorous common language for scientific modelling, analogous to the type theories of computer science. As a case study, we sketch how to translate from compositional narrative concepts to neural circuits and back again.

As machine learning black boxes are increasingly being deployed in critical domains such as healthcare and criminal justice, there has been a growing emphasis on developing techniques for explaining these black boxes in a human interpretable manner. It has recently become apparent that a high-fidelity explanation of a black box ML model may not accurately reflect the biases in the black box. As a consequence, explanations have the potential to mislead human users into trusting a problematic black box. In this work, we rigorously explore the notion of misleading explanations and how they influence user trust in black box models. More specifically, we propose a novel theoretical framework for understanding and generating misleading explanations, and carry out a user study with domain experts to demonstrate how these explanations can be used to mislead users. Our work is the first to empirically establish how user trust in black box models can be manipulated via misleading explanations.

Rank aggregation aims at producing a single ranking from a co llection of rankings of a fixed set of alternatives. In social choice theory (e.g., Moulin 1991), where the alternatives are candidates to an election and each ranking represents the preferences o f a voter, aggregation rules are called Social Welfare Functions (SWFs). Apart from social choice, rank aggregation has prov ed useful in many applications, including preference learning (Cheng a nd H ullermeier, 2009; Cl emen con et al., 2018), collaborative filtering (Wang et al., 2014), genetic map creation (Jackson et al., 2008), similarity search in databases systems (Fagin et al., 2003) and design of web search engines (Altman and Tennenholtz, 2008; Dwork et al., 2001). In the fo llowing, we use interchangeably the terms "input rankings" and "preferences", "output rank ing" and "consensus ranking", as well as "alternatives" and "'candidates". The well-known Arrow's impossibility theorem states that t here exists no aggregation rule satisfying a small set of desirable properties (Arrow, 1950). In the absense of an "ideal" rule, various aggregation rules have been proposed and studied. F ollowing Fishburn's classification (1977), we can distinguish between the SWFs for which the out put ranking can be computed from the majority graph alone, those for which the output ranking can be computed fro m the 1 Table 1: Results of setwise contests in Example 1. set c

We focus on the scenario in which messages pro and/or against one or multiple candidates are spread through a social network in order to affect the votes of the receivers. Several results are known in the literature when the manipulator can make seeding by buying influencers. In this paper, instead, we assume the set of influencers and their messages to be given, and we ask whether a manipulator ( e.g., the platform) can alter the outcome of the election by adding or removing edges in the social network. We study a wide range of cases distinguishing for the number of candidates or for the kind of messages spread over the network. We provide a positive result, showing that, except for trivial cases, manipulation is not affordable, the optimization problem being hard even if the manipulator has an unlimited budget ( i.e., he can add or remove as many edges as desired). Furthermore, we prove that our hardness results still hold in a reoptimization variant, where the manipulator already knows an optimal solution to the problem and needs to compute a new solution once a local modification occurs ( e.g., in bandit scenarios where estimations related to random variables change over time). Introduction Nowadays, social network media are the most used, if not the unique, sources of information. This indisputable fact turned out to influence most of our daily actions, and also to have severe effects on the political life of our countries. Indeed, in many of the recent political elections around the world, there has been evidence of the impact that false or incomplete news spread through these media influenced the electoral outcome. For example, in the recent US presidential election, Allcott and Gentzkow (2017) and Guess, Nyhan, and Reifler (2018) show that, on average, 92% of Americans remembered pro-Trump false news, while 23% of them remembered the pro-Clinton fake news. As another example, Ferrara (2017) shows that automated accounts in Twitter spread a considerable amount of political news in order to alter the outcome of 2017 French elections. In this scenario, a natural question is to understand at which extent the spread of (mis)information on social network media may alter the result of a political election. This topic has recently received large interest in the community: e.g., Auletta et al. (2015; 2017a; 2017b) show that, in the case of two only candidates, a manipulator may be able to lead the minority to become a majority by influencing the order in which voters change their mind.

Graph convolution is the core of most Graph Neural Networks (GNNs) and usually approximated by message passing between direct (one-hop) neighbors. In this work, we remove the restriction of using only the direct neighbors by introducing a powerful, yet spatially localized graph convolution: Graph diffusion convolution (GDC). GDC leverages generalized graph diffusion, examples of which are the heat kernel and personalized PageRank. It alleviates the problem of noisy and often arbitrarily defined edges in real graphs. We show that GDC is closely related to spectral-based models and thus combines the strengths of both spatial (message passing) and spectral methods. We demonstrate that replacing message passing with graph diffusion convolution consistently leads to significant performance improvements across a wide range of models on both supervised and unsupervised tasks and a variety of datasets. Furthermore, GDC is not limited to GNNs but can trivially be combined with any graph-based model or algorithm (e.g. spectral clustering) without requiring any changes to the latter or affecting its computational complexity. Our implementation is available online.

In healthcare, the highest risk individuals for morbidity and mortality are rarely those with the greatest modifiable risk. By contrast, many machine learning formulations implicitly attend to the highest risk individuals. We focus on this problem in point processes, a popular modeling technique for the analysis of the temporal event sequences in electronic health records (EHR) data with applications in risk stratification and risk score systems. We show that optimization of the log-likelihood function also gives disproportionate attention to high risk individuals and leads to poor prediction results for low risk individuals compared to ones at high risk. We characterize the problem and propose an adjusted log-likelihood formulation as a new objective for point processes. We demonstrate the benefits of our method in simulations and in EHR data of patients admitted to the critical care unit for intracerebral hemorrhage.

The Linley Fall Processor Conference was held on October 23 - 24 at the Hyatt Regency Hotel in Santa Clara, CA. Proceedings are now available at the link above. Click here to view the keynotes on our Youtube channel. The conference presentations featured AI acceleration, targeting edge, automotive, IoT, and data center. Also covered were new CPU architectures, networking, memory, security, SoC design, and other processor-related topics.

One of the world's leading brands in delivering digital experiences, Sitecore brings nearly two decades of expertise in reimagining customer experiences. From its founding in 2001 in Copenhagen, Denmark, the firm has become a powerhouse, known and relied on by its customers for its innovation and ability to push boundaries in technology to create differentiated digital experiences. Given that Sitecore operates in digital mediums, the company routinely handles vast arrays of data and content. Finding ways to maximize the value of that data and content has been an important focus for the company over the years. That is why the advent of artificial intelligence (AI) was of great interest to its leadership, for its potential to treat data and content in entirely new ways.

With opening of Doctor Sleep this past weekend, we are reminded of the visionary genius of Stanley Kubrick. His powerful impact on art, technology and, yes, even humanity itself remains. Perhaps his most profound contribution was his science-fiction masterpiece 2001: A Space Odyssey, which was produced in 1968 and was based on the novel (and previously, the short story "The Sentinel") by Arthur C. Clarke. The film explores human evolution, existentialism and, notably, artificial intelligence. This is pure visual poetry about the relationship between man and technology. In many ways, this piece of art inspired today's technology, which is now inspiring today's "art."

In my last blog we focussed on some of the problems with Artificial Intelligence (AI) and public trust that can be compounded by organisational issues such as dark data. This time round we're going to look at a couple of examples that demonstrate how AI can be used as a force for good. Over the past few months we have been working with the World Economic Forum (WEF) to test out some of the guidance on AI that we have been drafting with them. There have been a lot of lively debates as the use of AI is clearly divisive, especially when it comes to image processing. If we look at the UK there has been controversy recently over police using facial recognition techniques on CCTV footage to support the fight against crime.