Projects / Programmes
Causalify - Causality in global social dynamics
| Code |
Science |
Field |
Subfield |
| 2.07.07 |
Engineering sciences and technologies |
Computer science and informatics |
Intelligent systems - software |
| Code |
Science |
Field |
| P176 |
Natural sciences and mathematics |
Artificial intelligence |
| Code |
Science |
Field |
| 1.02 |
Natural Sciences |
Computer and information sciences |
artificial intelligence, causal modelling, probabilistic reasoning, language understanding, low carbon economy, knowledge graphs
Researchers (7)
| no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
| 1. |
22278 |
PhD Janez Brank |
Computer science and informatics |
Researcher |
2019 - 2022 |
95 |
| 2. |
28015 |
PhD Blaž Fortuna |
Computer science and informatics |
Researcher |
2019 - 2022 |
152 |
| 3. |
31885 |
PhD Aljaž Košmerlj |
Computer science and informatics |
Researcher |
2019 - 2022 |
34 |
| 4. |
33425 |
PhD Jurij Leskovec |
Computer science and informatics |
Researcher |
2019 - 2022 |
281 |
| 5. |
12570 |
PhD Dunja Mladenić |
Computer science and informatics |
Head |
2019 - 2022 |
662 |
| 6. |
34646 |
PhD Inna Novalija |
Computer science and informatics |
Researcher |
2019 - 2022 |
65 |
| 7. |
32381 |
PhD Primož Škraba |
Mathematics |
Researcher |
2019 - 2022 |
133 |
Organisations (1)
| no. |
Code |
Research organisation |
City |
Registration number |
No. of publicationsNo. of publications |
| 1. |
0106 |
Jožef Stefan Institute |
Ljubljana |
5051606000 |
90,682 |
Abstract
The next wave of Artificial Intelligence will be centred on extracting deeper structure from observed systems. Built on the rapid development of existing AI technologies, topics such as Causal Reasoning, Common Sense Reasoning, Text Understanding will result in much more powerful solutions than are possible today.
In Causalify, we target scenarios where we (a) monitor multiple complementary global real-time data streams, (b) interconnect them into an evolving probabilistic causal knowledge graph, (c) prepare an operational algorithmic platform, (d) answer and explain complex proactive and reactive questions about the world’s known and possibly unknown phenomena, (e) address ethical and human protective issues related to such up-coming AI technologies, and (f) apply the developed methods to the domain of the low carbon economy.
Our cross-disciplinary use-case is the field of global Low Carbon Economy, addressing societal, political, economic, health, food and environmental aspects. We have ensured access to multiple data streams covering different aspects of the relevant global dynamics in near real-time including media, market, supply chain, jobs & skills, science, weather and satellite images.
Sample questions we aim to answer are: predicting trends and follow-up events in the world, answering ‘what-if’ questions, causally explaining global phenomena in a human understandable way, predicting disruptions along supply chains, predicting the impact of science on the job market, and proactively spotting anomalies from earth observation data.
Methodologically, our starting points are the state-of-the-art AI approaches and tools. We expect key scientific results in theory, methodology and algorithms, especially in the fields of scalable multiresolution causality modelling, text understanding, scalable probabilistic reasoning, evolving knowledge graph construction from diverse data sources and new insights into the Low Carbon Economy.
Significance for science
The importance of the proposed project for the broad advancement of research is in addressing topics which will be most likely future topics of Artificial Intelligence, beyond existing “narrow AI tasks”. This includes deeper structural insights into observed systems like causality modelling, common sense reasoning, using state-of-the art machine learning (deep learning) to learn highly efficient reasoning operators, and text understanding. All these approaches will be combined to gain global insights into low carbon economy on large real data streams on global dynamics. An additional confirmation of the importance of the proposed research is the recent (Dec 2018) DARPA call on “Machine Common Sense” and upcoming call (mid 2019) on “Knowledge-directed Artificial Intelligence Reasoning Over Schemas” both addressing the topics of the proposed project Causalify.
Significance for the country
The importance of the proposed project for the broad advancement of research is in addressing topics which will be most likely future topics of Artificial Intelligence, beyond existing “narrow AI tasks”. This includes deeper structural insights into observed systems like causality modelling, common sense reasoning, using state-of-the art machine learning (deep learning) to learn highly efficient reasoning operators, and text understanding. All these approaches will be combined to gain global insights into low carbon economy on large real data streams on global dynamics. An additional confirmation of the importance of the proposed research is the recent (Dec 2018) DARPA call on “Machine Common Sense” and upcoming call (mid 2019) on “Knowledge-directed Artificial Intelligence Reasoning Over Schemas” both addressing the topics of the proposed project Causalify.
Most important scientific results
Interim report
Most important socioeconomically and culturally relevant results
