Agnese Chiatti
Research Fellow at Politecnico di Milano (Italy).
Knowledge Engineering for Robots,
Neuro-symbolic AI,
Computer Vision,
Service and Field Robotics
Research Fellow at Politecnico di Milano (Italy).
Knowledge Engineering for Robots,
Neuro-symbolic AI,
Computer Vision,
Service and Field Robotics
My project proposal for the MSCA Postdoctoral Fellowship “ReFiNe - Redefining Field robot deployment through Neuro-symbolic visual sensemaking” has received the EU Seal of Excellence (SoE). Currently funded by Politecnico di Milano through the SoE Fellowship, this project investigates how to assess the trustworthiness of Vision Language models in complex scenarios that require advanced visual capabilities through the use of graph knowledge representations.
I have contributed to AgrifoodTEF, a 60M euro EU funded initiative (2023-2028) involving partners from 9 EU countries aimed at providing the first EU-wide Testing and Experimentation Facility for the AI and Robotics solutions in Precision Agriculture such as autonomous weeding, harvesting, and precise plant spraying.
Awardee of the L’Oréal-UNESCO for Women in Science Fellowship to conduct research on visually intelligent robotic applications for precision agriculture.
I have contributed to the EU flagship project GATEKEEPER, which has focused on developing smart digital technologies to support health independent living for the ageing population. In this context, I have developed robot vision and semantic mapping solutions to retrieve personal items in home environments as an assistive tool in cases of mild cognitive and visual impairment.
My PhD research has been focused on enhancing the Visual Intelligence of service robots by combining Knowledge-driven technologies with Machine Learning. Specifically, I have developed a prototype of robot assistant which can monitor office environments in search for potentially dangerous situations (e.g., flammable items left by ignition sources, cluttered emergency exits, dangling cables, and others).
I have contributed to the EU-funded project SPICE (Social cohesion, participation, and Inclusion through Cultural Engagement), where I have explored the use of Deep Learning and Neurosymbolic Learning methods to classify artistic subjects from cultural heritage image collections.
I have been part of the organising team of the 1st Smart Cities and Robotics Challenge (SciRoc), which was held in Milton Keynes (UK) on September 2019. This was the first robotic challenge to be held in a public shopping mall, to explore the integration of robots within Smart City infrastructures.
While working in Prof. Lee Giles’ Lab at PSU, I have been a part of the Human Screenome project, a collaboration between Penn State University and Stanford University. I have been in charge of implementing an end-to-end architecture for extracting and indexing textual information from smartphone and laptop screenshots. This platform has allowed researchers in the behavioural and medical sciences to analyse how daily media consumption may affect fragile user categories, such as adolescents and low-income groups. Outcomes from this research have been featured in The New York Times, SAGE Ocean, and Medium, among others.
For an updated list, you can refer to my Google Scholar.
Catalano, N., Maranelli, A., Chiatti, A., and Matteucci, M.
More than the Sum of Its Parts: Ensembling Backbone Networks for Few-Shot Segmentation.
In Proceedings of the International Joint Conference on Neural Networks (IJCNN24).
Chiatti, A., Bertoglio, R., Catalano, N., Gatti, M., and Matteucci, M.
Surgical fine-tuning for Grape Bunch Segmentation under Visual Domain Shifts.
In Proceedings of the European Conference on Mobile Robots (ECMR23).
Chiatti, A., Bardaro, G., Matteucci, M., and Motta, E.
Visual Model Building for Robot Sensemaking: Perspectives, Challenges, and Opportunities.
Bridge Session on AI and Robotics of the thirty-seventh AAAI conference on Artificial Intelligence (AAAI-23).
Chiatti, A.
Visually Intelligent Agents: Improving Sensemaking in Service Robotics.
PhD Thesis. The Open University.
Chiatti, A., and Daga, E.
Neuro-symbolic learning for dealing with sparsity in
cultural heritage image archives: an empirical journey.
In Proceedings of the Workshop on Deep Learning for Knowledge Graphs at ISWC 2022 (DL4KG 2022). CEUR.
Chiatti, A., Bardaro, G., Motta, E., and Daga, E.
A Spatial Reasoning Framework for Commonsense Reasoning in Visually Intelligent Agents.
In Proceedings of the 8th International Workshop on Artificial Intelligence and Cognition (AIC 2022). CEUR.
Bardaro, G., Daga, E., Carvalho, J., Chiatti, A., and Motta, E.
Introducing a Smart City component in a Robotic Competition: a field report.
In Frontiers in Robotics and AI - Smart Sensor Networks and Autonomy. 9.
Chiatti, A., Motta, E., and Daga, E.
Robots with Commonsense: Improving Object Recognition through Size and Spatial Awareness.
In Proceedings of the 2022 AAAI Spring Symposium on Machine Learning and Knowledge Engineering for Hybrid Intelligence (AAAI-MAKE 2022). CEUR.
Chiatti, A.
Towards Visually Intelligent Agents (VIA): a Hybrid Approach.
In Proceedings of 2021 European Semantic Web Conference (ESWC 2021) Satellite events. Springer.
Chiatti, A., Motta, E., Daga, E., and Bardaro, G.
Fit to Measure: Reasoning about Sizes for Robust Object Recognition.
In Proceedings of the 2021 AAAI Spring Symposia - Workshop on Combining Machine Learning and Knowledge Engineering (AAAI-MAKE 2021). CEUR.
Chiatti, A., Motta, E., and Daga, E. Towards a Framework for Visual Intelligence in Service Robotics: Epistemic Requirements and Gap Analysis.
In Proceedings of the 17th International Conference on Principles of Knowledge Representation and Reasoning (KR 2020). IJCAI.
Chiatti, A., Bardaro, G., Bastianelli, E., Tiddi, I., Mitra, P., and Motta, E. Task-agnostic Object Recognition for Mobile Robots
through Few-shot Image Matching.
In Electronics. Special Issue on Big Data Analytics for Smart Cities. 9(3), 380. MDPI.
Reeves, B., Ram, N., Robinson, T.N., Cummings, J. J., Giles,L.,Pan,J., Chiatti, A., Cho, M.J. et al.
Screenomics: A Framework to Capture and Analyze Personal Life Experiences and the Ways that Technology Shapes Them.
In Human Computer Interaction.
Chiatti, A., Bardaro, G., Bastianelli, E., Tiddi, I., Mitra, P., and Motta, E. Exploring Task-agnostic, ShapeNet-based Object Recognition for Mobile Robots.
In Proceedings of the 3rd International workshop on Data Analytics solutions for Real-LIfe APplications (DARLI-AP). CEUR.
Ul Hoque, M.R., Bradley, D., Kwan, C., Chiatti, A., Li, J. and Wu, J.
Searching for Evidence of Scientific News in Scholarly Big Data..
In Proceedings of the 10th International Conference on Knowledge Capture (K-CAP 2019). ACM.
Bardaro, G., Semprebon, A., Chiatti, A., and Matteucci, M.
From Models To Software Through Automatic Transformations: An AADL To ROS End-to-End Toolchain..
In Proceedings of the Third IEEE International Conference on Robotic Computing (IRC), 580-585. IEEE.
Chiatti, A., Cho, M.J., Gagneja, A., Yang, X., Brinberg, M., Roehrick, K., Choudhury, S.R., Ram, N., Reeves, B. and Giles, C.L.
Text Extraction and Retrieval from Smartphone Screenshots: Building a Repository
for Life in Media. .
In Proceedings of the 33rd ACM/SIGAPP Symposium on
Applied Computing (SAC 2018). Pau, France. April 9-13, 2018. ACM.
Chiatti, A., Yang, X., Brinberg, M., Cho, M.J., Anupriya, A., Ram, N., Reeves, B. and Giles, C.L.
Text Extraction from Smartphone Screenshots to Archive in situ Media Behavior. .
In Proceedings of the 9th International Conference on Knowledge Capture (K-CAP
2017). Austin, TX, USA. December 4-6, 2017. ACM.
Wu, J. Choudhury, S.R., Chiatti, A., Liang, C, and Giles, C.L.
HESDK: A Hybrid Approach to Extracting Scientific Domain Knowledge Entities..
In Proceedings of the ACM/IEEE Joint Conference on Digital Libraries (JCDL 2017). 241-
244. Toronto, ON, Canada, June 19-23, 2017.
Chiatti, A., Dragisic, Z., Cerquitelli, T., and Lambrix, P.
Reducing the search space in ontology alignment using clustering techniques and topic
identification..
In Proceedings of the 8th International Conference on Knowledge
Capture (K-CAP 2015). Palisades, NY, USA, October 7-10, 2015. ACM.