Philipp Plank
Papers
1
Total Citations
5
H-Index
1
About
Philipp Plank is a researcher at the forefront of energy-efficient deep learning and neuromorphic computing. His work centers on bridging the gap between novel hardware architectures and advanced neural network models, with a particular focus on state space models (SSMs) and their deployment on specialized platforms like Intel’s Loihi 2. Plank’s major contribution lies in demonstrating how diagonal structured state space models can be efficiently mapped onto neuromorphic hardware, achieving remarkable energy savings while maintaining high performance for streaming sequence processing tasks. This work, already garnering early citations, addresses the critical challenge of unsustainable energy costs in modern AI systems. By showing that SSMs can run orders of magnitude more efficiently than traditional transformers on conventional GPUs, Plank is helping to chart a path toward more sustainable AI. His research is vital for students and researchers interested in the intersection of algorithm design, hardware-software co-optimization, and the future of low-power, real-time intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1