Spain

The amount of satellite imagery data has been substantially growing since the start of the SENTINEL mission. However, applications of AI to EO data are still scarce. The main goal of the AiTLAS project is to facilitate the uptake of EO data by AI experts and vice versa - the uptake of (advanced) AI methods by EO experts. The main focus of AiTLAS is the development of a comprehensive toolbox with resources such as: benchmarking tools, ready-to-exploit models, tools for learning models de novo, and semantically annotated datasets prepared in a format that is easy to use by AI methods. In this talk, we will illustrate the early developments of the AiTLAS project on a case study of learning deep models from satellite images scarcely labelled with ground-based land-use data. In particular, using large amounts of unlabelled satellite images, we will first learn (pre-train) in an unsupervised manner a deep model capable of extracting visual features, and then use a substantially smaller amount of labelled data for fine-tuning the model in a supervised manner. We will show that this approach has a strong competitive advantage as compared to using only labeled data for supervised learning of deep models.

ECAI 2020

AiTLAS: A toolbox of AI methods tailored for Earth Observation data

ECAI2020 featured a full programme of technical papers, as well as many AI-related events of interest for researchers, students and all attendants who are interested on contemporary AI, including:




**An extensive programme of workshops, tutorials, and invited speakers**

**The Prestigious Applications of Intelligent Systems conference (PAIS)**

**The Frontiers in AI track sessions**

**The Lab To Market event**

**The Women in AI time**

**The EU Challenges forum**

… and many more!

ECAI 2020 featured a full programme of technical papers, as well as many AI-related events of interest for researchers, students and all attendants who are interested on contemporary AI.

technical paper

This tutorial aims to provide an initial foundation in the concepts underlying computational argumentation and its role and relevance in the context of Human-Centric AI. We will also discuss potential applications in this context.

Computational Argumentation in the context of Human-Agent Interaction

The transition towards an electricity system with a growing presence of non-programmable Renewable Energy Sources (RESs) systems requires not only the implementation of modern control systems and equipment, able to manage bidirectional power flows according to the Smart Grid paradigm, but also the adoption of novel distribution electrical system structures and/or new frameworks for the energy management and the electricity market scenarios, where resources considered up to now unable to provide ancillary services may participate as new flexible Distributed Energy Resources (DERs) that the power system requires, including non-programmable renewable sources, and demand. A Smart Grid integrates the actions of all actors connected to it, such as generators, suppliers, aggregators, consumers, in order to efficiently deliver sustainable, economic and secure electricity supplies. It uses information and communication technology on top of the electric network to improve the reliability, security, and efficiency (both economic and energy) of the electric system. To achieve this, it is also required the deployment of sensors and smart metering for collecting data across the grid from DERs and storage systems at any point of the network, especially integration at the consumer/prosumer level. Thus, it creates the opportunity for any consumer/prosumer on the grid to become an energy/power supplier and take part in energy trading mechanism. This is the key for consumers to be converted into “active energy citizens”. An essential role is played by Demand Response which denotes variations of the electric consumption by users in response to the energy price changes over the time, or in the presence of financial incentives and reliability signals.

An IoT Architecture for a Virtual Power Plant Aggregator

Frontiers in AI talk: Representation Learning and Synthesis for Generalized Planning

A central challenge to contemporary AI is to integrate learning and reasoning. The integration of learning and reasoning has been studied for decades already in the fields of statistical relational artificial intelligence and probabilistic programming. StarAI has focussed on unifying logic and probability, the two key frameworks for reasoning, and has extended this probabilistic logics machine learning principles.
I will argue that StarAI and Probabilistic Logics form an ideal basis for developing neuro-symbolic artificial intelligence techniques. Thus neuro-symbolic computation = StarAI + Neural Networks.
Many parallels will be drawn between these two fields and will be illustrated using the Deep Probabilistic Logic Programming language DeepProbLog.

From Statistical Relational to Neural Symbolic Artificial Intelligence 

Since the early days of AI, Knowledge Representation and Reasoning (KR) has pursued the goal of capturing human knowledge in forms that can be stored in machines and used for automated inference, enabling computers to draw conclusions analogous to the ones we humans draw from our own knowledge. Although some goals of the field’s original research agenda proved elusive, KR has delivered solutions to many central AI problems. In this talk, we will discuss a few selected areas where KR has proved very successful, if not necessarily at replicating the whole spectrum of human intelligence, certainly at achieving more intelligent information systems. For example, explicit knowledge captured in ontologies is a powerful tool to access information on the Web, and for facilitating the management of data that may be unstructured, heterogeneous, and incomplete. When combined with rules, it can be leveraged for solving hard problems under incomplete information. Beyond these success stories, we will also touch on some current AI challenges where we believe KR can play a central role: can explicit knowledge and automated reasoning help us make modern AI more transparent, safer, and more trustworthy?

Frontiers in AI talk: Knowledge and Reasoning for Intelligent Systems

AI on a Beautiful Island: Building the Future of Energy Networks

The pace at which AI theory is being delivered to application has accelerated in the last decade, creating some impressive value in some areas (e.g., health and legal informatics, manufacturing, supply chain management), but raising warning flags about trust and ethics.

Both the promise and the challenges are evident in the application of AI to automotives and automated autonomous systems, especially in the choice of technologies, tradeoffs in where intelligence is required (i.e., the autonomous system or the infrastructure), and the emerging role of explainable AI, both in improving the transparency, trust, and robustness of systems, and in informing social systems and regulators about how to confirm their safety.

We try to highlight salient aspects of these challenges, and provide some context for helping to manage the translation of AI theory to application.

Frontiers in AI talk: At the frontier of AI application: balancing technology push and application pull

One of the most significant developments in the area of design verification over the last three decade is the development of algorithmic methods for verifying temporal specification of finite-state designs. A frequent criticism against this approach, however, is that verification is done after significant resources have already been invested in the development of the design. Since designs invariably contains errors, verification simply becomes part of the debugging process. The critics argue that the desired goal is to use temporal specification in the design development process in order to guarantee the development of correct designs. This is called temporal synthesis. In this talk I will review 60 years of research on the temporal synthesis problem, describe the automata-theoretic approach developed to solve this problem, and describe both successes and failures of this research program.

The Siren Song of Temporal Synthesis

The combination of autonomous robots, artificial intelligence and the internet of things offers immense possibilities to improve healthcare and assistance in daily living activities. Human-robot interaction in this context poses very demanding research challenges to ensure reliable and situated communication, as well as compliance with the strict physical contact requirements involved, for example, in helping users to dress or feed. These assistive tasks require dexterous manipulation skills, which need to be easily taught to robots by non-experts, as well as customized according to user needs and preferences. In addition, such skills must be intrinsically safe to people, adaptable to changing conditions, and able to handle deformable materials like clothing. Some results of projects addressing these challenges, such as CLOTHILDE and SOCRATES, will be described. Assistive AI raises also fundamental ethical issues, many practical ones stemming from algorithmic decision-making conflicting with human freedom and dignity. Several institutions are developing regulations and standards, and many ethics education initiatives include contents on human-machine interaction and human dignity in assistive situations. In the talk, free materials to teach a university course on Ethics in Social Robotics and AI, based on the appeal of science fiction narrative, will be presented.

Assistive AI Research Challenges, Ethics and Science Fiction

In a dynamic modal logic a 'lie that p' (where p is a some proposition) is an action, that is interpreted as a state transformer relative to the proposition p. These states are pointed Kripke models encoding the uncertainty of agents about their beliefs, and their transformation results in updated beliefs. Lies can be about factual propositions but also about the beliefs of other agents. Deception can be given meaning in terms of protocols consisting of sequences of such actions, in view of realizing an epistemic goal. Agents can have many different roles. Two speaker perspectives are: (obs) an outside observer who is lying to an agent that is modelled in the system, and (ag) an agent who is lying to another agent, and where both are modelled in the system. Three addressee perspectives are: the *credulous* agent who believes everything that it is told (even at the price of inconsistency), the *skeptical* agent who only believes what it is told if that is consistent with its current beliefs, and the *belief revising* agent who believes everything that it is told by consistently revising its current, possibly conflicting, beliefs. Then again there may be non-addressed agents perceiving lies and deception yet different from the addressee. Lying may be costly. Not only in terms of delayed response time in psychological experiments, on which we will not speak, but also in terms of computational complexity of performing certain tasks while lying or deceiving, or while taking into account that one might be lied to. We do not know of hard results in this area but there seem to be many interesting open questions. Results on computational complexity seem to be particularly relevant for AI. Issues are not merely the possibility of one lie among many truths coming out of the mouth of a single agent (we recall Ulam games), but also the presence of one unreliable agent in the presence of many trustworthy agents (typical in security protocol settings). How to detect lies or liars, and how costly is that? There is ample room for special case studies and benchmarks, such as lying in gossip protocols. Background: Hans van Ditmarsch, Petra Hendriks, Rineke Verbrugge: Editors' Review and Introduction: Lying in Logic, Language, and Cognition. topiCS 12(2): 466-484 (2020)

AiTLAS: A toolbox of AI methods tailored for Earth Observation data

Downloads

Next from ECAI 2020

Computational Argumentation in the context of Human-Agent Interaction

Similar lecture

Cloud Removal from Satellite Imagery using Multispectral Edge-filtered Conditional Generative Adversarial Networks

AiTLAS: A toolbox of AI methods tailored for Earth Observation data

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ECAI 2020

Computational Argumentation in the context of Human-Agent Interaction

Similar lecture

Cloud Removal from Satellite Imagery using Multispectral Edge-filtered Conditional Generative Adversarial Networks

Downloads