Doctoral Fellow

**Job Description**
This project focuses on investigating event-driven learning approaches within Reinforcement Learning (RL) to enhance data efficiency through meta-learning and pre-training, facilitating few-shot adaptations. The research aims to demonstrate how the synergy between event-triggered learning and meta-learning can significantly increase RL efficiency at test time. An additional avenue explores the co-design of algorithms and digital neuromorphic hardware to further improve method efficiency, specifically investigating digital, event-based implementations of RL learning rules. Benchmarking will assess accuracy, latency, data efficiency, and energy consumption on small robotic control tasks.

**Skills & Abilities**
• Strong coding skills for programming neural networks, machine learning, and machine learning software frameworks (e.g., PyTorch or Jax).
• Ability for creative and analytical thinking across discipline boundaries and abstraction levels.
• Knowledge in integrated circuit design, testing, and simulation using Cadence (plus).
• Knowledge of digital neuromorphic hardware and sensors (plus).
• Ability for collaborative work, interdisciplinary and cross-topical thinking.
• Very good communication skills in English, both spoken and written.

**Qualifications**
Required Degree(s) in:
• Physics
• Electrical/Electronic Engineering
• Computer Science
• Mathematics
• A related field (Master’s degree)

Note: We’ve analyzed the actual job post using AI, for more details visit the original job post by clicking on “Apply Now”!