I am the Principal Investigator of the Creative Problem Solving in Cognitive Systems (CreaCogs) project funded by the German Research Foundation (DFG), and part of the Cognitive System (CoSy) group at the Bremen Spatial Cognition Center within FB3 (Faculty of Mathematics and Informatics) at Bremen Universität.
I work at the intersection of artificial intelligence and cognitive science, trying to understand how humans solve various tasks and build computational approaches which enable artificial intelligence systems to solve similar classes of tasks. My current focus is on creative problem solving.
My current project investigates representations and processes which can be applied in cognitively-inspired ArtificiaI Intelligence for creative problem-solving. My general interest is to enable creative problem-solving and creation of new useful knowledge in cognitive AI systems, and the study/modeling of human creativity.
Humans operate well in environments over which they hold incomplete knowledge, by using their previously acquired knowledge in new ways. Thus humans are creative when solving abstract and concrete problems, being able to draw on a variety of representational formats, make analogical transfers, make visuospatial inferences, create new concepts, new heuristics and new problem representations. I study the processes and types of knowledge organization which can enable such creative problem-solving, in artificial and natural cognitive systems.
As part of my current work, I use various computational tools to make cognitive systems which can answer creative problem solving tasks and/or help investigate how humans solve such tasks. These systems are designed under a theoretical framework for creative problem solving (CreaCogs). Here are some highlights:
- - comRAT-C is a system that can solve the compound Remote Associates Test
- - OROC is a prototype which can do creative object replacement and object composition
- - comRAT-V can solve a visual variant of the Remote Associates Test which we created
- - comRAT-G creates Remote Associates Test items, over the entirety of frequently used American English terms (17 million items created from noun terms alone). Work on comRAT-G aims to provide a larger repository of queries for cognitive psychologists and help them control their queries based on term frequency and answer probability.
In order to enable these systems I designed the theoretical framework and various computational approaches. The aim of some such approaches is to tackle a variety of creative problem solving tasks in an integrated computational manner – for example to have one framework the knowledge organization of which allows for a core set of creative inference processes to be deployed, which will allow for tasks to be solved. Other approaches aim to build the way to computationally understanding re-representation, classes of creative problem solving, etc.
I also perform empirical investigations (psychological experiments) to understand how humans solve such tasks and evaluate the computational cognitive systems built. I design new such tasks to investigate different parts of the creative problem solving processes. For example I designed a visual counterpart for the verbal Remote Associates Test. You can see our initial set of stimuli here, for a more advanced one please contact me. I also designed an approach for creating new insight problems in the practical objects domain, and others.
For a more complete publication list see https://www.researchgate.net/profile/Ana-Maria_Olteteanu
For access to some of our datasets, repositories of test items which we created, speaker invites or interest in undertaking an internship as part of the project, contact me directly.
For recent news on awards, papers and conference participation go here .