Research
My research lies at the intersection of machine learning, neuromorphic computing, adaptive intelligent systems, and computational ideas inspired by physics and neuroscience.
Spiking neural networks are one important part of this picture, but my broader interest is in how intelligence can emerge from temporal dynamics, sparse computation, local adaptation, structured memory, and hardware-aware constraints.
Core research
Spiking Neural Networks
- Deep and structured SNN architectures
- Temporal coding and sparse dynamics
- STDP and reward-modulated plasticity
- Online and continual adaptation
Neuromorphic Computing
- Event-driven computation
- Low-power and hardware-aware intelligence
- Learning for constrained systems
- Edge and energy-limited settings
Adaptive Intelligence
- Multi-timescale learning
- Fast adaptation and slow consolidation
- Memory, forgetting, and stability
- Learning under changing conditions
Broader interests
Machine Learning and Neural Networks
- Deep learning architectures
- Efficient and compact models
- Representation learning
- Learning with limited resources or supervision
Physics- and Neuroscience-Inspired Computation
- Dynamical systems perspectives on learning
- Biologically inspired learning rules
- Predictive and recurrent processing
- Structured memory and attractor-like mechanisms
Intelligent Systems under Constraints
- Energy-aware adaptation
- Sensing, computation, and communication trade-offs
- Learning in low-power environments
- System-level regulation of intelligence
Research practice
Research Software
- Modular experimental pipelines
- Reusable ML and SNN tooling
- Reproducible workflows
- Prototype systems for new learning ideas
Experimental Workflows
- Model design and ablation studies
- Implementation-oriented research
- Evaluation under realistic constraints
- Bridging theory and engineering practice
Current directions
deep and structured spiking architectures;
local and reward-based learning rules;
multi-timescale adaptation and consolidation;
predictive and memory-oriented computational frameworks;
adaptive intelligence under energy and resource constraints.
Perspective
I am especially interested in systems that are dynamic, sparse, adaptive, and physically grounded. In practice, this means working across different levels at once: conceptual questions, model design, experimental implementation, and software tooling.