Artificial intelligence seems to be taking over the world with systems that model pixels, words, and phonemes. The world is arguably made up, not of pixels, words, and phonemes but of entities (objects, things, including events) with properties and relations among them. Surely we should model these, not the perception or description of them. You might suspect that concentrating on modeling words and pixels is because all of the (valuable) data in the world is in terms of text and images. If you look into almost any company you will find their most valuable data is in spreadsheets, databases and other relational formats. These are not the form that are studied in introductory machine learning, but are full of product numbers, student numbers, transaction numbers and other identifiers that can't be interpreted naively as numbers. The field that studies this sort of data has various names including relational learning, statistical relational AI, and many others. This paper explains why relational learning is not taking over the world -- except in a few cases with restricted relations -- and what needs to be done to bring it to it's rightful prominence.

Relational databases are a successful model for data storage, and rely on query languages for information retrieval. Most of these query languages are based on relational algebra, a mathematical formalization at the core of relational models. Knowledge graphs are flexible data storage structures that allow for knowledge completion using machine learning techniques. Knowledge hypergraphs generalize knowledge graphs by allowing multi-argument relations. This work studies knowledge hypergraph completion through the lens of relational algebra and its core operations. We explore the space between relational algebra foundations and machine learning techniques for knowledge completion. We investigate whether such methods can capture high-level abstractions in terms of relational algebra operations. We propose a simple embedding-based model called Relational Algebra Embedding (ReAlE) that performs link prediction in knowledge hypergraphs. We show theoretically that ReAlE is fully expressive and can represent the relational algebra operations of renaming, projection, set union, selection, and set difference. We verify experimentally that ReAlE outperforms state-of-the-art models in knowledge hypergraph completion, and in representing each of these primitive relational algebra operations. For the latter experiment, we generate a synthetic knowledge hypergraph, for which we design an algorithm based on the Erdos-R'enyi model for generating random graphs.

Second-order Møller-Plesset perturbation theory using the Resolution-of-the-Identity approximation (RI-MP2) is a state-of-the-art approach to accurately estimate many-body electronic correlation effects. This is critical for predicting the physicochemical properties of complex molecular systems; however, the scale of these calculations is limited by their extremely high computational cost. In this paper, a novel many-GPU algorithm and implementation of a molecular-fragmentation-based RI-MP2 method are presented that enable correlated calculations on over 180,000 electrons and 45,000 atoms using up to the entire Summit supercomputer in 12 minutes. The implementation demonstrates remarkable speedups with respect to other current GPU and CPU codes, excellent strong scalability on Summit achieving 89.1% parallel efficiency on 4600 nodes, and shows nearly-ideal weak scaling up to 612 nodes. This work makes feasible ab initio correlated quantum chemistry calculations on significantly larger molecular scales than before on both large supercomputing systems and on commodity clusters, with a potential for major impact on progress in chemical, physical, biological and engineering sciences.

AIspace is a set of tools used to learn and teach fundamental AI algorithms. The original version of AIspace was written in Java. There was not a clean separation of the algorithms and visualization; it was too complicated for students to modify the underlying algorithms. Its next generation, AIspace2, is built on AIPython, open source Python code that is designed to be as close as possible to pseudocode. AISpace2, visualized in JupyterLab, keeps the simple Python code, and uses hooks in AIPython to allow visualization of the algorithms. This allows students to see and modify the high-level algorithms in Python, and to visualize the output in a graphical form, aiming to better help them to build confidence and comfort in AI concepts and algorithms. So far we have tools for search, constraint satisfaction problems (CSP), planning and Bayesian network. In this paper we outline the tools and give some evaluations based on user feedback.

Knowledge graphs store facts using relations between two entities. In this work, we address the question of link prediction in knowledge hypergraphs where relations are defined on any number of entities. While techniques exist (such as reification) that convert non-binary relations into binary ones, we show that current embedding-based methods for knowledge graph completion do not work well out of the box for knowledge graphs obtained through these techniques. To overcome this, we introduce HSimplE and HypE, two embedding-based methods that work directly with knowledge hypergraphs. In both models, the prediction is a function of the relation embedding, the entity embeddings and their corresponding positions in the relation. We also develop public datasets, benchmarks and baselines for hypergraph prediction and show experimentally that the proposed models are more effective than the baselines.

Landslides, movement of soil and rock under the influence of gravity, are common phenomena that cause significant human and economic losses every year. Experts use heterogeneous features such as slope, elevation, land cover, lithology, rock age, and rock family to predict landslides. To work with such features, we adapted convolutional neural networks to consider relative spatial information for the prediction task. Traditional filters in these networks either have a fixed orientation or are rotationally invariant. Intuitively, the filters should orient uphill, but there is not enough data to learn the concept of uphill; instead, it can be provided as prior knowledge. We propose a model called Locally Aligned Convolutional Neural Network, LACNN, that follows the ground surface at multiple scales to predict possible landslide occurrence for a single point. To validate our method, we created a standardized dataset of georeferenced images consisting of the heterogeneous features as inputs, and compared our method to several baselines, including linear regression, a neural network, and a convolutional network, using log-likelihood error and Receiver Operating Characteristic curves on the test set. Our model achieves 2-7% improvement in terms of accuracy and 2-15% boost in terms of log likelihood compared to the other proposed baselines.

Usage of Graphics Processing Units (GPU) has become strategic for simulating the chemistry of large molecular systems, with the majority of top supercomputers utilizing GPUs as their main source of computational horsepower. In this paper, a new fragmentation-based Hartree-Fock matrix build algorithm designed for scaling on many-GPU architectures is presented. The new algorithm uses a novel dynamic load balancing scheme based on a binned shell-pair container to distribute batches of significant shell quartets with the same code path to different GPUs. This maximizes computational throughput and load balancing, and eliminates GPU thread divergence due to integral screening. Additionally, the code uses a novel Fock digestion algorithm to contract electron repulsion integrals into the Fock matrix, which exploits all forms of permutational symmetry and eliminates thread synchronization requirements. The implementation demonstrates excellent scalability on the Summit computer, achieving good strong scaling performance up to 4096 nodes, and linear weak scaling up to 612 nodes.

Knowledge graphs are used to represent relational information in terms of triples. To enable learning about domains, embedding models, such as tensor factorization models, can be used to make predictions of new triples. Often there is background taxonomic information (in terms of subclasses and subproperties) that should also be taken into account. We show that existing fully expressive (a.k.a. universal) models cannot provably respect subclass and subproperty information. We show that minimal modifications to an existing knowledge graph completion method enables injection of taxonomic information. Moreover, we prove that our model is fully expressive, assuming a lower-bound on the size of the embeddings. Experimental results on public knowledge graphs show that despite its simplicity our approach is surprisingly effective.

Statistical relational AI (StarAI) aims at reasoning and learning in noisy domains described in terms of objects and relationships by combining probability with first-order logic. With huge advances in deep learning in the current years, combining deep networks with first-order logic has been the focus of several recent studies. Many of the existing attempts, however, only focus on relations and ignore object properties. The attempts that do consider object properties are limited in terms of modelling power or scalability. In this paper, we develop relational neural networks (RelNNs) by adding hidden layers to relational logistic regression (the relational counterpart of logistic regression). We learn latent properties for objects both directly and through general rules. Back-propagation is used for training these models. A modular, layer-wise architecture facilitates utilizing the techniques developed within deep learning community to our architecture. Initial experiments on eight tasks over three real-world datasets show that RelNNs are promising models for relational learning.

Knowledge graphs contain knowledge about the world and provide a structured representation of this knowledge. Current knowledge graphs contain only a small subset of what is true in the world. Link prediction approaches aim at predicting new links for a knowledge graph given the existing links among the entities. Tensor factorization approaches have proved promising for such link prediction problems. Proposed in 1927, Canonical Polyadic (CP) decomposition is among the first tensor factorization approaches. CP generally performs poorly for link prediction as it learns two independent embedding vectors for each entity, whereas they are really tied. We present a simple enhancement of CP (which we call SimplE) to allow the two embeddings of each entity to be learned dependently. The complexity of SimplE grows linearly with the size of embeddings. The embeddings learned through SimplE are interpretable, and certain types of background knowledge can be incorporated into these embeddings through weight tying. We prove SimplE is fully expressive and derive a bound on the size of its embeddings for full expressivity. We show empirically that, despite its simplicity, SimplE outperforms several state-of-the-art tensor factorization techniques. SimplE's code is available on GitHub at https://github.com/Mehran-k/SimplE.

Statistical relational models provide compact encodings of probabilistic dependencies in relational domains, but result in highly intractable graphical models. The goal of lifted inference is to carry out probabilistic inference without needing to reason about each individual separately, by instead treating exchangeable, undistinguished objects as a whole. In this paper, we study the domain recursion inference rule, which, despite its central role in early theoretical results on domain-lifted inference, has later been believed redundant. We show that this rule is more powerful than expected, and in fact significantly extends the range of models for which lifted inference runs in time polynomial in the number of individuals in the domain. This includes an open problem called S4, the symmetric transitivity model, and a first-order logic encoding of the birthday paradox. We further identify new classes S2FO2 and S2RU of domain-liftable theories, which respectively subsume FO2 and recursively unary theories, the largest classes of domain-liftable theories known so far, and show that using domain recursion can achieve exponential speedup even in theories that cannot fully be lifted with the existing set of inference rules.

We consider the problem of, given a probabilistic model on a set of random variables, how to add a new variable that depends on the other variables, without changing the original distribution. In particular, we consider relational models (such as Markov logic networks (MLNs)), where we cannot directly define conditional probabilities. In relational models, there may be an unbounded number of parents in the grounding, and conditional distributions need to be defined in terms of aggregators. The question we ask is whether and when it is possible to represent conditional probabilities at all in various relational models. Some aggregators have been shown to be representable by MLNs, by adding auxiliary variables; however it was unknown whether they could be defined without auxiliary variables. For other aggregators, it was not known whether they can be represented by MLNs at all. We obtained surprisingly strong negative results on the capability of flexible undirected relational models such as MLNs to represent aggregators without affecting the original model's distribution. We provide a map of what aspects of the models, including the use of auxiliary variables and quantifiers, result in the ability to represent various aggregators. In addition, we provide proof techniques which can be used to facilitate future theoretic results on relational models, and demonstrate them on relational logistic regression (RLR).

Various representations and inference methods have been proposed for lifted probabilistic inference in relational models. Many of these methods choose an order to eliminate (or branch on) the parameterized random variables. Similar to such methods for non-relational probabilistic inference, the order of elimination has a significant role in the performance of the algorithms. Since finding the best order is NP-complete even for non-relational models, heuristics have been proposed to find good orderings in the non-relational models. In this paper, we show that these heuristics are inefficient for relational models, because they fail to consider the population sizes associated with logical variables. We extend existing heuristics for non-relational models and propose new heuristics for relational models. We evaluate the existing and new heuristics on a range of generated relational graphs.

The drive toward sustainable wastewater management is challenging the conventional paradigm of linear end-of-pipe solutions. A shift toward more sustainable solutions requires that information about new ideas, systems, and technologies be more readily accessible for addressing wastewater problems. It is commonly argued that decision-making needs to involve engineers and other community representatives to define values and brainstorm solutions. This paper describes a decision support system (DSS) prototype that is designed to help community planners identify solutions which balance environmental, economic, and social goals. The system is designed to be scalable, adaptable, and flexible to allow fair assessment of new ideas and technologies. It supports the exploration of consequences of various alternatives and visualizes the tradeoffs between them. Our DSS takes in modular descriptions of components and a description of a community context, automates the design of alternative wastewater systems, and facilitates evaluating how well each design satisfies the given context. It provides an adaptable platform from which new solutions can be designed without having to predefine how a single component fits within a specific system. Our DSS facilitates the exploration of alternative solutions by visualizing the effect of various tradeoffs and their consequences in relation to the community's sustainability goals.

Abstract. The aim of what semantic science is to have scientific ontologies, data, and hypotheses represented and published in machine understandable forms that enable predictions on new cases. There is much work on developing scientific ontologies and representing scientific data in terms of these ontologies. The next step is to publish hypotheses that can make (probabilistic) predictions on the published data and can be used for prediction on new cases. The published data can be used to evaluate hypotheses. To make a prediction in a particular case, hypotheses are combined to form models. This paper considers feature-based semantic science where the data and new cases are described in terms of features. A prediction for a new case is made by building a model made up of hypotheses that fit together, are consistent with the ontologies used, and are adequate for the case. We give some desiderata for such models, and show how the construction of such models is a form of abduction. We provide a definition for models that satisfies these criteria and prove that it produces a coherent probability distribution over the values of interest.

Inductive Logic Programming (ILP) is an area of Machine Learning which has now reached its twentieth year. Using the analogy of a human biography this paper recalls the development of the subject from its infancy through childhood and teenage years. We show how in each phase ILP has been characterised by an attempt to extend theory and implementations in tandem with the development of novel and challenging real-world applications. Lastly, by projection we suggest directions for research which will help the subject coming of age.

In environmental and natural resource planning domains actions are taken at a large number of locations over multiple time periods. These problems have enormous state and action spaces, spatial correlation between actions, uncertainty and complex utility models. We present an approach for modeling these planning problems as factored Markov decision processes. The reward model can contain local and global components as well as spatial constraints between locations. The transition dynamics can be provided by existing simulators developed by domain experts. We propose a landscape policy defined as the equilibrium distribution of a Markov chain built from many locally-parameterized policies. This policy is optimized using a policy gradient algorithm. Experiments using a forestry simulator demonstrate the algorithm's ability to devise policies for sustainable harvest planning of a forest.

Interactive algorithm visualizations (AVs) are powerful tools for teaching and learning concepts that are difficult to describe with static media alone. However, while countless AVs exist, their widespread adoption by the academic community has not occurred due to usability problems and mixed results of pedagogical effectiveness reported in the AV and education literature. This paper presents our experiences designing and evaluating CIspace, a set of interactive AVs for demonstrating fundamental Artificial Intelligence algorithms. In particular, we first review related work on AVs and theories of learning. Then, from this literature, we extract and compile a taxonomy of goals for designing interactive AVs that address key pedagogical and usability limitations of existing AVs. We advocate that differentiating between goals and design features that implement these goals will help designers of AVs make more informed choices, especially considering the abundance of often conflicting and inconsistent design recommendations in the AV literature. We also describe and present the results of a range of evaluations that we have conducted on CIspace that include semi-formal usability studies, usability surveys from actual students using CIspace as a course resource, and formal user studies designed to assess the pedagogical effectiveness of CIspace in terms of both knowledge gain and user preference. Our main results show that (i) studying with our interactive AVs is at least as effective at increasing student knowledge as studying with carefully designed paper-based materials; (ii) students like using our interactive AVs more than studying with the paper-based materials; (iii) students use both our interactive AVs and paper-based materials in practice although they are divided when forced to choose between them; (iv) students find our interactive AVs generally easy to use and useful. From these results, we conclude that while interactive AVs may not be universally preferred by students, it is beneficial to offer a variety of learning media to students to accommodate individual learning preferences. We hope that our experiences will be informative for other developers of interactive AVs, and encourage educators to exploit these potentially powerful resources in classrooms and other learning environments.

This paper describes the design of the CIspace interactive visualization tools for teaching and learning Artificial Intelligence. Our approach to design is to iterate through three phases: identifying pedagogical and usability goals for supporting both educators and students, designing to achieve these goals, and then evaluating our system. We believe identifying these goals is essential in confronting the usability deficiencies and mixed results about the pedagogical effectiveness of interactive visualizations reported in the Education literature. The CIspace tools have been used and positively received in undergraduate and graduate classrooms at the University of British Columbia and internationally. We hope that our experiences can inform other developers of interactive visualizations and encourage their use in classrooms and other learning environments.

Mark-recapture models have for many years been used to estimate the unknown sizes of animal and bird populations. In this article we adapt a finite mixture mark-recapture model in order to estimate the number of active telephone lines in the USA. The idea is to use the calling patterns of lines that are observed on the long distance network to estimate the number of lines that do not appear on the network. We present a Bayesian approach and use Markov chain Monte Carlo methods to obtain inference from the posterior distributions of the model parameters. At the state level, our results are in fairly good agreement with recent published reports on line counts. For lines that are easily classified as business or residence, the estimates have low variance. When the classification is unknown, the variability increases considerably. Results are insensitive to changes in the prior distributions. We discuss the significant computational and data mining challenges caused by the scale of the data, approximately 350 million call-detail records per day observed over a number of weeks.

Many artificial intelligence (AI) tasks, such as product configuration, decision support, and the construction of autonomous agents, involve a process of constrained optimization, that is, optimization of behavior or choices subject to given constraints. In this paper we present an approach for constrained optimization based on a set of hard constraints and a preference ordering represented using a CP‐network—a graphical model for representing qualitative preference information. This approach offers both pragmatic and computational advantages. First, it provides a convenient and intuitive tool for specifying the problem, and in particular, the decision maker's preferences. Second, it admits an algorithm for finding the most preferred feasible (Pareto‐optimal) outcomes that has the following anytime property: the set of preferred feasible outcomes are enumerated without backtracking. In particular, the first feasible solution generated by this algorithm is Pareto optimal.

Information about user preferences plays a key role in automated decision making. In many domains it is desirable to assess such preferences in a qualitative rather than quantitative way. In this paper, we propose a qualitative graphical representation of preferences that reflects conditional dependence and independence of preference statements under a ceteris paribus (all else being equal) interpretation. Such a representation is often compact and arguably quite natural in many circumstances. We provide a formal semantics for this model, and describe how the structure of the network can be exploited in several inference tasks, such as determining whether one outcome dominates (is preferred to) another, ordering a set outcomes according to the preference relation, and constructing the best outcome subject to available evidence.

This paper shows how state-of-the-art state estimation techniques can be used to provide efficient solutions to the difficult problem of real-time diagnosis in mobile robots. The power of the adopted estimation techniques resides in our ability to combine particle filters with classical algorithms, such as Kalman filters. We demonstrate these techniques in two scenarios: a mobile waiter robot and planetary rovers designed by NASA for Mars exploration.

Current recommender systems, based on collaborative filtering, implement a rather limited model of interaction. These systems intelligently elicit information from a user only during the initial registration phase. Furthermore, users tend to collaborate only indirectly. We believe there are several unexplored opportunities in which information can be effectively elicited from users by making the underlying interaction model more conversational and collaborative. In this paper, we propose a set of techniques to intelligently select what information to elicit from the user in situations in which the user may be particularly motivated to provide such information. We argue that the resulting interaction improves the user experience. We conclude by reporting results of an offline experiment in which we compare the influence of different elicitation techniques on both the accuracy of the systems predictions and the users effort

Bayesian belief networks have grown to prominence because they provide compact representations for many problems for which probabilistic inference is appropriate, and there are algorithms to exploit this compactness. The next step is to allow compact representations of the conditional probabilities of a variable given its parents. In this paper we present such a representation that exploits contextual independence in terms of parent contexts; which variables act as parents may depend on the value of other variables. The internal representation is in terms of contextual factors (confactors) that is simply a pair of a context and a table. The algorithm, contextual variable elimination, is based on the standard variable elimination algorithm that eliminates the non-query variables in turn, but when eliminating a variable, the tables that need to be multiplied can depend on the context. This algorithm reduces to standard variable elimination when there is no contextual independence structure to exploit. We show how this can be much more efficient than variable elimination when there is structure to exploit. We explain why this new method can exploit more structure than previous methods for structured belief network inference and an analogous algorithm that uses trees.

This paper discusses the application of particle filtering algorithms to fault diagnosis in complex industrial processes. We consider two ubiq- uitous processes: an industrial dryer and a level tank. For these appli- cations, we compared three particle filtering variants: standard parti- cle filtering, Rao-Blackwellised particle filtering and a version of Rao- Blackwellised particle filtering that does one-step look-ahead to select good sampling regions. We show that the overhead of the extra process- ing per particle of the more sophisticated methods is more than compen- sated by the decrease in error and variance.

A new method is proposed for exploiting causal independencies in exact Bayesian network inference. A Bayesian network can be viewed as representing a factorization of a joint probability into the multiplication of a set of conditional probabilities. We present a notion of causal independence that enables one to further factorize the conditional probabilities into a combination of even smaller factors and consequently obtain a finer-grain factorization of the joint probability. The new formulation of causal independence lets us specify the conditional probability of a variable given its parents in terms of an associative and commutative operator, such as ``or'', ``sum'' or ``max'', on the contribution of each parent. We start with a simple algorithm VE for Bayesian network inference that, given evidence and a query variable, uses the factorization to find the posterior distribution of the query. We show how this algorithm can be extended to exploit causal independence. Empirical studies, based on the CPCS networks for medical diagnosis, show that this method is more efficient than previous methods and allows for inference in larger networks than previous algorithms.

As influence diagrams become a popular representational tool for decision analysis, influence diagram evaluation attracts more and more research interests. In this article, we present a new, two‐phase method for influence diagram evaluation. In our method, an influence diagram is first mapped onto a decision graph and then the analysis is carried out by evaluating the decision graph. Our method is more efficient than Howard and Matheson's because, among other reasons, our method generates a much smaller decision graph for the same influence diagram. Like those most recent algorithms reported in the literature, our method also provides a clean interface between influence diagram evaluation and Bayesian net evaluation. Consequently, various well‐established algorithms for Bayesian net evaluation can be used in influence diagram evaluation. Furthermore, our method has a few unique merits. First, it takes advantage of asymmetry in influence diagrams to avoid unnecessary computation. Second, by using heuristic search techniques, it provides an explicit mechanism for using heuristic information that may be available in a domain‐specific form. These additional merits make our method more efficient than the current algorithms in general. Finally, by using decision graphs as an intermediate representation, the value of perfect information can be computed in a more efficient way.

Bayesian networks provide a natural, concise knowledge representation method for building knowledge‐based systems under uncertainty. We consider domains representable by general but sparse networks and characterized by incremental evidence where the probabilistic knowledge can be captured once and used for multiple cases. Current Bayesian net representations do not consider structure in the domain and lump all variables into ahomogeneousnetwork. In practice, one often directs attention to only part of the network within a period of time; i.e., there is “localization” of queries and evidence. In such case, propagating evidence through a homogeneous network is inefficient since the entire network has to be updated each time. This paper derives reasonable constraints, which can often be easily satisfied, that enable a natural{localization preserving)partition of a domain and its representation by separate Bayesian subnets. The subnets are transformed into a set of permanent junction trees such that evidential reasoning takes place at only one of them at a time; and marginal probabilities obtained are identical to those that would be obtained from the homogeneous network. We show how to swap in a new junction tree, and absorb previously acquired evidence. Although the overall system can be large, computational requirements are governed by the size of one junction tree.

This article investigates two different activities that involve making assumptions: predicting what one expects to be true and explaining observations. In a companion article, a logic-based architecture for both prediction and explanation is proposed and an implementation is outlined. In this article, we show how such a hypothetical reasoning system can be used to solve recognition, diagnostic, and prediction problems. As part of this is the assumption that the default reasoner must be “programmed” to get the right answer and it is not just a matter of “stating what is true” and hoping the system will magically find the right answer. A number of distinctions have been found in practice to be important: between predicting whether something is expected to be true versus explaining why it is true; and between conventional defaults (assumptions as a communication convention), normality defaults (assumed for expediency), and conjectures (assumed only if there is evidence). the effects of these distinctions on recognition and prediction problems are presented. Examples from a running system are given.

Although there are many arguments that logic is an appropriate tool for artificial intelligence, there has been a perceived problem with the monotonicity of classical logic. This paper elaborates on the idea that reasoning should be viewed as theory formation where logic tells us the consequences of our assumptions. The two activities of predicting what is expected to be true and explaining observations are considered in a simple theory formation framework. Properties of each activity are discussed, along with a number of proposals as to what should be predicted or accepted as reasonable explanations. An architecture is proposed to combine explanation and prediction into one coherent framework. Algorithms used to implement the system as well as examples from a running implementation are given.