Lay summarisation aims to produce summaries of scientific articles that are comprehensible to non-expert audiences. However, previous work assumes a one-size-fits-all approach, where the content and style of the produced summary are entirely dependent on the data used to train the model. In practice, audiences with different levels of expertise will have specific needs, impacting what content should appear in a lay summary and how it should be presented. Aiming to address this, we propose ATLAS, a novel abstractive summarisation approach that can control various properties that contribute to the overall "layness" of the generated summary using targeted control attributes. We evaluate ATLAS on a combination of biomedical lay summarisation datasets, where it outperforms state-of-the-art baselines using mainstream summarisation metrics. Additional analyses provided on the discriminatory power and emergent influence of our selected controllable attributes further attest to the effectiveness of our approach.

I teach managing technological innovation in Simon Fraser University's Management of Technology MBA program. No matter our industry or field, we should regularly review our tools and workflows. New tools, like AI, are excellent triggers for this assessment. Sorting out how best to adjust our work, as per the values and existing norms of different fields, takes a systematic approach. My research examines how companies can adjust how they use talent, technology and technique to hit work targets and stay aligned with the times -- what I've called thinking in 5T.

Organisations must conduct end-to-end audits that consider both the social and technical aspects of artificial intelligence (AI) to fully understand the impacts of any given system, but a lack of understanding around how to conduct holistic audits and the limitations of the process is holding back progress, say algorithmic auditing experts. At the inaugural International Algorithmic Auditing Conference, hosted in Barcelona on 8 November by algorithmic auditing firm Eticas, experts had a wide-ranging discussion on what a "socio-technical" audit for AI should entail, as well as various challenges associated with the process. Attended by representatives from industry, academia and the third sector, the goal of the conference is to create a shared forum for experts to discuss developments in the field and help establish a roadmap for how organisations can manage their AI systems responsibly. Those involved in this first-of-its-kind gathering will go on to Brussels to meet with European Union (EU) officials and other representatives from digital rights organisations, so they can share their collective thinking on how AI audits can and should be regulated for. Gemma Galdon-Clavell, conference chair and director of Eticas, said: "Technical systems, when they're based on personal data, are not just technical, they are socio-technical, because the data comes from comes from social processes."

The SPATIAL (Security and Privacy Accountable Technology Innovations, Algorithms, and machine Learning) project seeks to address the challenges of black-box AI and data management in cybersecurity by designing and developing resilient accountable metrics, privacy-preserving methods, verification tools and system framework that will serve as critical building blocks to achieve trustworthy AI in security solutions. In addition to this, the project aims to help generate appropriate skills and education for trustworthy AI in cybersecurity on both societal and technical aspects. The project covers data privacy, resilience engineering, and legal-ethical accountability that are in line with EU top agenda to achieve trustworthy AI. In addition, the work carried out in SPATIAL on both social and technical aspects will serve as a stepping stone to establish an appropriate governance and regulatory framework for AI-driven security in Europe.

In this blog post, I'll be introducing you to Capsule Networks (CapsNet), along with highlighting its significance in the domain of deep learning. Before diving into CapsNet, let's take a look at CNN's and their limitations, which CapsNet addressed. Convolutional neural networks (CNNs) are one of the reasons why deep learning has become so popular in recent years. CNN's were created to map image data to a variable output. They've proven to be so successful that they're now the method of choice for any form of prediction problem utilizing image data as an input.

This paper attempts to synthesize various conceptualizations of the term "context" as found in computing literature. Ten conceptual dimensions of context thus emerge -- location; user, task, and system characteristics; physical, social, organizational, and cultural environments; time-related aspects, and historical information. Together, the ten dimensions of context provide a comprehensive view of the notion of context, and allow for a more systematic examination of the influence of context and contextual information on human-system or human-AI interactions.

With availability of massive data and computation, Machine Learning (ML) and other spheres of artificial intelligence are growing at rapid rate. AI has become the demand of time and the need of the hour. To keep up, almost every company is either starting a new Data Science/Machine Learning department or expanding rapidly with multiple projects in pipeline. Now, we have more ML competitions and hackathons than ever recorded in the history.Everyday there are new courses focusing entirely on Python libraries and Machine Learning APIs. People are sharing latest machine learning (ML) algorithms, computations, graphs, charts and code snippets on a daily basis focusing technical aspects and implementations.

In the field of IP law, however, AI raises new questions and challenges. A research project of the legal departments of the Max Planck Institute for Innovation and Competition led by Professor Reto M. Hilty and Professor Josef Drexl is investigating these issues. The Research Group "Regulation of the Digital Economy" examines whether the existing IP system can fulfil its fundamental functions in the context of AI. Since a sound understanding of technology is indispensable for this task, the members of the group researched technical literature, conducted interviews with practitioners and organized a workshop with international AI researchers. The result is the present paper "Technical Aspects of Artificial Intelligence: An Understanding from an Intellectual Property Law Perspective".

Word embeddings are basically a form of word representation that bridges the human understanding of language to that of a machine. Word embeddings are distributed representations of text in an n-dimensional space. These are essential for solving most NLP problems. Domain adaptation is a technique that allows Machine learning and Transfer Learning models to map niche datasets that are all written in the same language but are still linguistically different. For example, legal documents, customer survey responses, and news articles are all unique datasets that need to be analyzed differently.

The whirl of reinforcement learning started with the advent of AlphaGo by DeepMind, the AI system built to play the game Go. Since then, various companies have invested a great deal of time, energy, and research, and today reinforcement learning is one of the hot topics within Deep Learning. That said, most businesses are struggling to find use cases for reinforcement learning or ways to encompass it within their business logic. So far, it's been studied only in risk-free, observed, environments that are easy to simulate, which means that industries like finance, health, insurance, tech-consultancies are reluctant to risk their own money to explore its applications. What's more, the aspect of "risk factoring" within reinforcement learning puts a high strain on systems.