Scholar
Rishabh Tiwari
Google Scholar ID: iJuoc4sAAAAJ
CS PhD @ UC Berkeley
Citations & Impact
Citations
508
H-index
7
i10-index
6
Publications
15
Co-authors
26
list available
Publications
Resume (English only)
Academic Achievements
- Publications: 'QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache' (Under Review, 2025), 'Using Early Readouts to Mediate Featural Bias in Distillation' (WACV 2024), 'Overcoming Simplicity Bias in Deep Networks using a Feature Sieve' (ICML 2023), 'GCR: Gradient Coreset Based Replay Buffer Selection For Continual Learning' (CVPR 2022), 'Chipnet: Budget-aware pruning with heaviside continuous approximations' (ICLR 2021).
Research Experience
- Work Experience: Pre-Doctoral Researcher at Google Deepmind. Research Projects: Developing algorithms to improve group-robustness of networks; mitigating simplicity bias; optimizing network architecture by developing novel network pruning algorithms. Positions: Ph.D. Student at UC Berkeley, Senior Researcher at Transmute AI Labs.
Education
- Degree: Ph.D. School: University of California, Berkeley (UC Berkeley). Advisor: Prof. Kurt Keutzer. Time: Present. Major: Electrical Engineering and Computer Sciences (EECS).
Background
- Research Interests: Efficient deep learning. Field: Computer Science and Engineering. Bio: Currently a Ph.D. student at BAIR, UC Berkeley, focusing on developing algorithms to improve group-robustness of networks and mitigate simplicity bias. Also a founding member and senior researcher at Transmute AI Labs.
Miscellany
- Personal Interests: Guiding undergraduate students in research. Contact: Email (rishabhtiwari at berkeley dot edu), Google Scholar page, LinkedIn page.
