|
196 | 196 | education: "B.E. in Electrical and Computer Engineering, Thapar Institute of Engineering and Technology, Patiala, India" |
197 | 197 | active: 1 |
198 | 198 | projects: |
199 | | - -title: "Consolidate and advance the GPU infrastructure in Clad" |
200 | | - status: Ongoing |
201 | | - description: | |
202 | | - Automatic Differentiation(AD) is a set of techniques to evaluate the derivative |
203 | | - of functions specified by the computer programs precisely and efficiently. Clad |
204 | | - is a Clang based AD tool that transforms C++ code to compute derivatives. Clad |
205 | | - supports multiple differentiation modes like reverse mode, forward mode and hessian |
206 | | - mode making it suitable for wide range of applications. Clad have also demonstrated |
207 | | - promising support for GPU based differentiation however the work remains fragmented. |
208 | | - This project aims to consolidate the fragmented work and advance Clad's GPU infrastructure |
209 | | - into a robust and consistent system. |
210 | | - proposal: /assets/docs/Vedant_Goyal_Proposal_2026.pdf |
211 | | - mentors: Aaron Jomy, David Lange, Vassil Vassilev |
| 199 | + - title: "Consolidate and advance the GPU infrastructure in Clad" |
| 200 | + status: Ongoing |
| 201 | + description: | |
| 202 | + Automatic Differentiation(AD) is a set of techniques to evaluate the derivative |
| 203 | + of functions specified by the computer programs precisely and efficiently. Clad |
| 204 | + is a Clang based AD tool that transforms C++ code to compute derivatives. Clad |
| 205 | + supports multiple differentiation modes like reverse mode, forward mode and hessian |
| 206 | + mode making it suitable for wide range of applications. Clad have also demonstrated |
| 207 | + promising support for GPU based differentiation however the work remains fragmented. |
| 208 | + This project aims to consolidate the fragmented work and advance Clad's GPU infrastructure |
| 209 | + into a robust and consistent system. |
| 210 | + proposal: /assets/docs/Vedant_Goyal_Proposal_2026.pdf |
| 211 | + mentors: Aaron Jomy, David Lange, Vassil Vassilev |
212 | 212 |
|
213 | 213 | - name: Matthew Barton |
214 | 214 | info: "Open Source Contributor" |
|
0 commit comments