Spain

In this paper, we identify a series of problems to track and understand what AI is capable of, surveying some previous initiatives. We present the AIcollaboratory, a data-driven framework to collect and explore data about AI results, progress and ultimately capabilities, being developed in the context of AI Watch, the European Commission (EC) knowledge service to monitor the development, uptake and impact of AI in Europe. We close the paper with some challenges for the community emerging around the AIcollaboratory.

ECAI 2020

Tracking AI: The Capability is (Not) Near

ai watch

aicollaboratory

ai progress

ai evaluation

The AGI conference series is the only major conference series devoted wholly and specifically to the creation of AI systems possessing general intelligence at the human level and ultimately beyond. By gathering together active researchers in the field, for presentation of results and discussion of ideas, we accelerate our progress toward our common goal.

Underline hosted the two main days of AGI 2020 on Wednesday June 24th and Thursday June 25th. Both days were **live-streamed** and you can read the full program bellow. All presentations are available for a **asynchronous viewing**. All presenters are grouped into panels. You can also hear guest talks by **Ben Goertzel** (SingularityNET), **Valery Tarasov** (Semenov Institute of Chemical Physics), **Tatiana Shavrina** (Sberbank) and **Joscha Bach** (MIT Media Lab). 

AGI-20

The AGI conference series is the only major conference series devoted wholly and specifically to the creation of AI systems possessing general intelligence at the human level and ultimately beyond. By gathering together active researchers in the field, for presentation of results and discussion of ideas, we accelerate our progress toward our common goal.

poster

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.

Ontologies are widely used to formally represent abstract domain knowledge. Logic reasoning ensures the logical consistency of ontologies, and infers knowledge implicitly encoded in ontologies. It has been shown both theoretically and empirically that for large and complex ontologies, reasoning is still time-consuming and resource-intensive. Meta-reasoning exploits machine learning techniques to tackle the important problems of understanding the source of reasoning hardness and to predict reasoning efficiency, with the overall goal of improving reasoning efficiency. In this paper, we highlight recent advances in meta-reasoning for Semantic Web ontologies, briefly present technical innovations and results, and discuss important problems for future research.

Towards Meta-reasoning for Ontologies: A Roadmap

The WikiTalk open-domain dialogue system performs Wikipedia-based talking. This paper highlights the potential for a new Wikipedia-based listening capability. The aim is to enable social robots to follow the changing topics in human-human conversations, opening the possibility for robots to join in human conversations by  making topically relevant contributions at appropriate points.

WikiTalk and WikiListen: Towards listening robots that can join in conversations with topically relevant contributions

Recent advances in language model (LM) pre-training from large- scale corpora have shown to improve various natural language processing tasks. They achieve performances comparable to non-expert humans on the GLUE benchmark for natural language understanding (NLU). While the improvement of the different contextualized representations comes from (i) the usage of more and more data, (ii) changing the types of lexical pre-training tasks or (iii) increasing the model size, NLU is more than memorizing word co-occurrences. But how much world knowledge and common sense can those language model capture? How much can those models infer from just the lexical information? To overcome this problem, some approaches include semantic information in the training process. We highlight existing approaches to combine different types of semantics with language models during the pre-training or fine-tuning phase

Enriching Language Models with Semantics

Lifted algorithms use representatives for groups of indistinguishable objects to efficiently perform inference. Standard lifted algorithms like first-order variable elimination or first-order knowledge compilation, compute answers to marginal queries of single random variables or events in a lifted way using representatives. But, queries containing a set of indistinguishable random variables may lead to groundings, something that lifting tries to avoid. This paper presents parameterised queries as a means to avoid groundings, applying the lifting idea to queries. Parameterised queries enable lifted algorithms to compute answers faster, while compactly representing queries and answers.

Lifting Queries for Lifted Inference

Agent-Based Models (ABMs) can be used to numerically simulate highly non-linear phenomena that emerge from local interactions of multiple, independent entities. ABMs are often used to understand dynamic processes such as tissue regeneration in biomedicine, in which group-level patterns and space-time constraints are of interest. High fidelity ABMs, especially those in biomedical applications, usually have a large number of parameters, creating substantial uncertainty and high dimensionality at both local and global levels. Uncertainty
 analysis (output variance estimation) and sensitivity analysis are essential steps in investigating model robustness. In addition to allocating uncertainty to each parameter, sensitivity analysis can also be used to reduce ABM dimensionality for effective model calibration and optimization. I discuss common sensitivity analysis methods that have been used to decrease the number of parameters in complex ABMs, highlighting Garg et al. (2019)’s paper, where random forests – a non-parametric ensemble algorithm – are used as a sensitivity analysis method for ranking parameters in a biomedical ABM.

Sensitivity Analysis for Dimensionality Reduction in Agent-Based Modeling

A picture is worth a thousand words”, the adage reads. However, pictures cannot replace words in terms of their ability to efficiently convey clear (mostly) unambiguous and concise knowledge. Images and text, indeed, reveal different and complementary information that, if combined, result in more information than the
 sum of that contained in the single media. The combination of visual and textual information can be obtained by linking the entities mentioned in the text with those shown in the pictures. To further integrate this with agent background knowledge, an additional step is necessary. That is, either finding the entities in the agent knowledge base that correspond to those mentioned in the text or shown in the
 picture or, extending the knowledge base with the newly discovered entities. We call this complex task Visual-Textual-Knowledge Entity Linking (VTKEL). In this paper, we precisely define the VTKEL task and present two datasets composed of 1k and 30k pictures, annotated with visual and textual entities, and linked to the YAGO ontology. Successively, we develop the first unsupervised algorithm for the solution of the VTKEL task. The evaluation of the algorithm shows promising results on both 1k and 30k VTKEL datasets.

VT-LINKER: Visual-Textual-Knowledge Entity Linker

In this paper, we present an approach for generating dialogues for characters within the context of computational narratives using personality-based features for deep neural networks. The approach integrates the requirements of both narrative genres and personality traits for the definition of character-based stylistic models. The modelling of characters’ features from existing datasets of complete stories permits the generation of personality-rich character dialogues. We present early results from an evaluation based on a sample of characters’ personality traits across different narrative genres, demonstrating variability in the resulting dialogues.

Tracking AI: The Capability is (Not) Near

Downloads

Next from ECAI 2020

Towards Meta-reasoning for Ontologies: A Roadmap

Similar lecture

Analyzing Elementary School Olympiad Math Tasks as a Benchmark for AGI