News
We are thrilled to announce that our contribution to this year’s ICML Topological Deep Learning Challenge secured both 2nd and 3rd place in the competition’s respective categories!
Topological Deep Learning enhances neural network performance by leveraging topological structures in data. Our approach centered around a hypergraph-based solution to optimize information flow in large-scale networks. Hypergraphs have proven effective in diverse applications, including medical knowledge graph optimization, logistics, and business process enhancement.
The 2024 Topological Deep Learning Challenge was jointly organized by TAG-DS and PyT-Team and hosted by the Geometry-grounded Representation Learning and Generative Modeling (GRaM) Workshop at ICML 2024.
We’re proud of our team’s success and look forward to continuing to innovate in the field of topological deep learning.
More information about the challenge and its outcomes can be found here.
Event
Join us at Conf42 Cloud Native 2025, where Perelyn’s Maksymilian Kalek will share practical strategies for building scalable AWS applications on a budget—without compromising performance.
Event
Perelyn attended E-world 2025 to explore AI’s impact on the energy sector. Engaging with industry experts, we discussed innovations, challenges, and AI-driven solutions for a smarter, more sustainable future.
