ABSTRACT. What if you could reduce your planning process from one week to one hour, or from one hour to one second? What if you could, at the click of a button, improve your bottom line by double digits? In this session, you will learn to do just that by leveraging machine learning (ML) and Decision Optimization (DO) technologies together. You will learn the differences and complementary strengths of ML and DO, learn about best practices, and see examples of combining these technologies to achieve financial gains and efficiencies. The session includes demos covering use cases such as marketing campaign planning and predictive maintenance.

ABSTRACT. Boxever have been providing 1:1 customer personalisation capabilities to some of the worlds top airlines and financial services businesses for over 6 years. Engage our customer engagement engine was built on Drools and had a coding interface which allowed coding of rules in a user friendly domain specific language created by Boxever to provide real time personalisation As our customers matured however they demanded the ability to arbitrate between marketing messages, service messages and customers service messages. They also were seeking ways to apply AI and machine learning across all customer touch points. This case study will discuss the challenges we faced both from a technical and user experience perspective in creating a decisioning capability based on the DMN standard for first best action and next best action and how these were overcome. The key takeaways will be: Building the right DMN modeller for enterprise decisioning – augmenting the DMN standard to improve flexibility Choosing the right scalable architecture to support multi-tenant enterprise deployment Decisioning and collections - To contain or not to contain Testing Decision Models - Component, Model and Variant Analytics - Choosing the right KPI's Challenges of building optimisation workflows (A/B/N) into model deployment

Audience: Business/Technical

Speakers:

Roy Robinson Director of Product, Boxever Roy has spent the past 10 years as a product specialist working with SaaS businesses to build out data and analytics platforms. He has a wealth of experience in building scalable real time decisioning products which leverage the latest data science techniques in both the security and customer experience fields. Roy has a passion for user experience and practical applications of AI. Roy has a Masters in Mathematics and Computer Science from Queens University, Belfast and is a SciFi and Fantasy novel fanatic!

Alan Giles CTO, Boxever Alan is a Co-Founder and the CTO at Boxever. He is a highly accomplished software technologist, software visionary, and entrepreneur with over 14 years’ experience leading, developing and delivering highly scalable and fault tolerant distributed systems. He is passionate about large-scale systems, cloud security, and building new architectures to support petabyte-scale data processing. In addition to developing world-class technology for the travel industry, Alan is a mentor and advisor in the Dublin startup scene and you can even find him occasionally at an all-night hackathon.

ABSTRACT. Artistic creativity in artificial intelligence is a phenomenon that indicates the self-organizing systems (SOS), through combinational, exploratory and transformational features. My research on computational artworks and the computer programs that generate those artworks, aims to introduce that the function of artificial intelligence is far beyond being merely a tool to create art, it is rather an actor that has an artistic and creative agency. By exploring the creative aspects in artificial intelligence in the domain of art, and reflecting on what is creativity and how it works cognitively in human & machine intelligence, I would like to present A.I. Art (or computational art) as an autonomous genre that redefines the dynamics between the artwork, the art-maker and the audience.

ABSTRACT. Self-Organising Systems have been a subject of research in various disciplines such as Biology, Cybernetics or Sociology. In this talk, I will provide an overview of the different approaches and discuss example applications in engineered (computer) systems including network design, load balancing and others.

ABSTRACT. In 2014 we introduced the "The Decision Model" method by B. v. Halle and L. Goldberg to our business analysts and launched the OpenRules rule engine in IT. At the same time, we could win a first project to use both the method and the new rule engine for the implementation of underwriting decisions in the individual life insurance. In 2017 we switched from "The Decision Model" to DMN and introduced the tool Innovator for Decision Modeling. After a total of more than four years of experience in decision modeling, we should have developed hundreds of Decisions by now. In fact, to date there have been about 4,000 rules in 54 decisions. What were the reasons why we haven't already turned all business logic into decisions? Many requirements must be met for the use of decisions to be accepted in both business and IT. The business must be ready to take responsibility for the business logic again and to manage it. Business analysts must be affine with the decision method and have appropriate tools to support them in their work. IT must be ready to leave the business logic back to the business. However, there are already many business applications that contain their own rule components and in which the business logic is partly maintained by IT. The transition from the modeled decision in Business to the executable decision in IT must take place without additional effort if possible. For this purpose, we have implemented a model-driven approach from the outset. And finally, the tools used in business and IT must not require too high investments, at least at the beginning. This presentation describes, • how we introduced a method for the presentation of rules in Business for the first time with "The Decision Model" and started with a cost-effective solution in initial projects, • switched to the OMG standard DMN in 2017 and used Innovator for decision modeling in Business, • which hurdles are to be expected when using Decisions.

Key take-away: Challenges in application of decision models in Business and IT Audience: business, new Industry Sector: Insurance Key Technology: DMN, Innovator, OpenRules

ABSTRACT. Waiting times for patients in Irish hospitals have grown in recent years to unprecedented levels, at the beginning of 2018 there were just over 500,000 people waiting to see a specialist for an outpatient appointment, of a total population of 4.77 million in Ireland. We investigate the use of data analytics and optimization to understand the current situation, predict future short and medium term demand, and study the impact of changes in processes and resource levels. While overall resource levels are clearly inadequate, we also study the impact of did-not-attend no-shows, the possibility of overbooking, and of load distribution across clinics and hospitals. Some of the analysis is based on publicly available data, with results made available in at https://patientwaitlistanalysis.wordpress.com/

ABSTRACT. Recent developments in AI, Machine Learning and Robotics have raised concerns about the ethical consequences of both academic and industrial AI research. Leading academics, businessmen and politicians have voiced an increasing number of questions about the consequences of AI not only over people, but also on the large-scale consequences on the the future of work and employment, its social consequences and the sustainability of the planet. In this work, we analyse the use and the occurrence of ethics-related research in leading AI, machine learning and robotics venues. In order to do so we perform long term, historical corpus-based analyses on a large number of flagship conferences and journals. Our experiments identify the prominence of ethics-related terms in published papers and presents several statistics on related topics. Finally, this research provides quantitative evidence on the pressing ethical concerns of the AI community.

ABSTRACT. Replaceability is a form of generalized substitutability whose features make it potentially of great importance for problem simplification. It differs from simple substitutability in that it only requires that substitutable values exist for every solution containing a given value without requiring that the former always be the same. This is also the most general form of substitutability that allows inferences from local to global versions of this property. Building on earlier work, this study first establishes that algorithms for localized replaceability (called consistent neighbourhood replaceability or CNR algorithms) based on all-solutions neighbourhood search outperform other replaceability algorithms by several orders of magnitude. It also examines the relative effectiveness of different forms of depth-first CNR algorithms. Secondly, it demonstrates an apparent complexity ridge, which does not occur at the same place in the problem space as the complexity areas for consistency or full search algorithms. Thirdly, it continues the study of methods for inferring replaceability in structured problems in order to improve efficiency. This includes correcting an oversight in earlier work and extending the formal analysis. It is also shown that some strategies for inferring replaceable values can be extended to disjunctive constraints in scheduling problems.

ABSTRACT. With the current surge of interest in ethics in AI, we present our position with respect to these challenges. Our proposal, responsible technologies, aims to (1) address a number of the ethical challenges put forward in AI, and (2) provide the first building blocks towards the development of ethical AI systems. The current discussion on how to address ethics in AI usually focuses on issues like policies, education, or research culture. There is no computational method yet mature enough to address ethics in AI. We break ground by proposing new methods and tools, underpinned by multidisciplinary research, that can make humans and machines understand their respective dynamic goals while strictly abiding by the values that inspire our societies. This position paper presents our plan of work for the development of responsible technologies that embed values within technology through what we refer to as ethics by construction.

ABSTRACT. After a brief overview of digital artwork constructed round games of life, a digital artwork entitled 'Traces' will be presented with under-the-hook considerations on some basic learning and evolutionary principles as well as artistic reflection on the digital traces we disseminate on the web and their potential implication for other to forecast future plausible personal evolution scenarii.

ABSTRACT. The liquid organization is a mode of organization of a non-standard artistic practice that cannot be captured by the institutional organizations mode, which is rigid by definition. This mode of organization is at the same time a survival necessity for the practice and a constraint for its development. The liquid organization composes with the environment. The water changes state but not nature.

ABSTRACT. Product Rule Systems are well known for being at the heart of Decision Magement, however the paradigm, with all its pitfalls, has not changed that much over the decades. They typically offer a monolithic design with a single working memory and rulebase, with limited rule orchestration and very quickly hit the limitations with rule chaining complexity. Approaches like DMN take a different strategy and simplify things by reducing the capabilities down to those needed for the specific types of problems DMN aims to solve. However there still remains a need for more powerful systems too, but they must modernise if they are to retain industry relevance.

This talk will discuss the ongoing innovations that are helping to modernise Product Rule Systems, while keep within the context of the Drools rule engine. Those innovations involve new techniques for rule modularisation, rule orchestration, recursion control and argumentation as well as breaking away from the traditional monolithic design of a single working memory and rule base.

ABSTRACT. The Climate Change Act and Social Value Act in the UK place responsibilities on many private sector and all public sector organisations to consider not only the value to customers and stakeholders of their activities but also the effect on greenhouse gases and the society they operate within.

The healthcare sector has a particularly significant place in this work, forming as it does a large part of the public sector and being present in every part of UK society. This programme that we've been involved with for some years now aims to recruit that considerable geographic & economic muscle to positively drive social and environmental action.

The programme collates the best and most up to date climate and social value modelling available and applies it to the problem of calculating a complete Sustainability report for each and every hospital trust and clinical commissioning group (CCGS - who commission primary healthcare providers) in England. In numerical terms this deals with almost 500 data inputs for each of around 200 hospital trusts, a further 200 CCGs as well as Ambulance Trusts, Mental Health providers and so on.

This case study will show how DMN has been used to document the previously black box system allowing far more people to understand, critique and improve the reporting process. Furthermore by making these decision models executable we have dramatically reduced the complexity of maintaining and updating the system as new and updated data becomes available.

ABSTRACT. Self-organization is a process where some form of overall order arises from local interactions between parts of an initially disordered system, this is spontaneous, not needing control. In my talk I will explain that there are also traces of self-organizing systems visible in contemporary artists' activities.

ABSTRACT. Aad Nales takes a step by step on how to build a robot that can autonomously move through its environment and that is sufficiently trained to take the right decisions. Dobots, the company he works for is part of Almende (english: the commons) a group of companies that make the support of self organisation its mission.

ABSTRACT. This industry track invites you to discuss the interaction between researchers, industry, AI and rules. We see that industry often need more simple solutions than offered by researchers but they need to be full proofed and that is much harder to achieve than a first prototype. But some processes in the industry are too difficult for people. People need to guess or 'don’t know': in those cases even very small improvements that increase forecasts from random (50%) to anything above 50% may already be a huge time and money saver for the industry. Researchers are often after bigger improvements. To apply a technology one needs common sense, domain expertise and a deep understanding of its working (benefits and limitations). When we talk about AI (the technology is an algorithm) and todays hype market, we expect to see many failures due to a lack of this insight. There is a challenge for researchers in the ruleML community to make sure people understand how AI works.

ABSTRACT. An example of a simple solutions that need to be full proofed is the rule language that is developed by the Dutch tax administration to express the rules for the income tax calcultation. Lawyers are being used to unconstrained natural and programmers are used to a procedural rule language. To process to get from one to the other was long and error prone. A new declarative language is replacing that process and allows the customer to gradually modernize their legacy calculation engines. This case shows again that the development of the rule language is just a very small thing in all steps for an organization to accept a new way of working.

ABSTRACT. An example of a simple solutions that need to be full proofed is the rule language that is developed by Isis Paparus to configure the workflow in their document management software. The business user defines general constraints in a controlled natural lanugage, such as "a customer must have an account manager" and "a claim must be assessed" to guide the workflow. The process of defining the language, limit the scope and supporting it in the user interface and training the customers shows the complexity of bringing a first prototype to production. Isis Papyrus also uses reinformcement learning to teach the software and the user what 'the next best step' in the process is based on past performance. This simple solution already works for years and is very popular.

ABSTRACT. Being a professional in IT, authority in business rules and specialist in artificial intelligence,  I have seen many initiatives in the industry to create  decision support systems and rule languages. Unfortunately many of them failed. This is not because the technology failed, but because there need to be many other business factors pointing in the right direction for a change initiative to succeed. We will discuss what research is needed to increase the succes factor of rule languages.

ABSTRACT. The financial services industry has long been a driver for technology innovation. Volatile regulations, increased demands for compliance, and a requirement for transparency necessitate the ability to quickly create and easily manage decisions across systems. The evolution of disruptive “FinTech” systems has attempted to address some of these needs. However, those requirements coupled with a constant demand for faster online processes has now created the need for Regulatory Technology or “RegTech”. A key component of new systems will certainly be the ability to create services or disseminate decisions in a consistent, unambiguous and transparent fashion.

The Mortgage Industry Standards and Maintenance Organization (MISMO) strives to allow participants in the mortgage industry (mortgage lenders, investors, servicers, industry vendors, borrowers) to exchange information and do so more securely, efficiently and economically. Although primarily focusing on an XML-based data standard thus far, they now seek to enable seamless exchange of all information between mortgage industry partners. To that end, MISMO is in the final stages of officially endorsing DMN as the recommended decision modeling standard for the mortgage industry.

This presentation will focus on the integration of business standards in the mortgage industry (MISMO) with technology standards (DMN, BPMN) to enable a powerful approach to handling FinTech and RegTech. These approaches will be demonstrated with their application to new mortgage requirements and regulations around the Uniform Residential Loan Application (URLA) and eMortgage. In addition, a new set of industry players will be empowered to succeed by having their compliance burden eased.