Kevin D. Murphy
Papers
1
Total Citations
21
H-Index
1
About
Kevin D. Murphy is a leading researcher in machine learning and probabilistic reasoning, with a focus on sequential decision-making and differentiable inference. His work bridges the gap between classical state estimation and modern deep learning, particularly through innovations in differentiable particle filters. His most-cited paper, "Towards Differentiable Resampling" (2020, 21 citations), tackles a fundamental challenge in end-to-end learning for recursive state estimation: making the resampling step—a core but non-differentiable component of particle filters—amenable to gradient-based optimization. This contribution has opened new avenues for integrating Bayesian filtering with neural networks, enabling more robust and adaptive models for time-series and control tasks. Murphy’s research is widely recognized for its clarity and practical impact, influencing both theoretical advances and applied systems in robotics, autonomous navigation, and sensor fusion. His work continues to shape how researchers approach probabilistic programming and differentiable inference, making him a key figure in the evolving landscape of modern machine learning.
Research Focus
Key Achievements
Top Papers
- 1Towards Differentiable Resampling21 citations · 2020