Chandan Kumar Jha

About Me

I am a Principal Research Scientist (साइंसदान्) at Indian Institute of Technology Bombay (IIT Bombay). I am a member of the Computer Architecture and Dependable Systems Lab (CADSL) led by Prof. Virendra Singh. I am working in the area of VLSI Design and Computer Architecture.

Education

PhD in Electrical Engineering, Indian Institute of Technology Gandhinagar (IIT Gandhinagar)
Thesis: “Approximate Designs and Frameworks for Error Resilient Applications”
(Advisor: Prof. Joycee Mekie)
CGPA : 9.16/10 (JUL 2015 - NOV 2020)

B.Tech in Electronics and Communcations Engineering, National Institute of Technology Meghalay (NIT Meghalaya)
CGPA : 9.38/10 (JUL 2011 - MAY 2015)

Internships

I worked with the SAFARI Research Group led by Prof. Onur Mutlu at ETH Zurich. The area of research was in-memory computing. (JUL 2019 - JAN 2020)

I worked with the Big Core Group at Intel Banagalore India. The area of research was On Chip Power Distribution Networks. (JAN 2017 - JUN 2017)

Publications

Conferences

  1. Neelam Sharma, Varun Venkitaraman, Newton Singh, Vikash Kumar, Shubham Singhania, and Chandan Kumar Jha, "Data-aware Cache Management for Graph Analytics," Accepted in Design, Automation and Test in Europe Conference 2022 (DATE '22). Accepted!!
  2. Chandan Kumar Jha, Sumit Walia, Gagan Kanojia, and Joycee Mekie, "FPCAM: Floating Point Configurable Approximate Multiplier for Error Resilient Applications," Accepted in IEEE International Symposium on Circuits and Systems 2021 (ISCAS '21). [Paper]
  3. Ankita Nandi, Chandan Kumar Jha, and Joycee Mekie, "Tunable Inexact Subtractors for Division in Image Processing Applications," Accepted in IEEE International Midwest Symposium on Circuits and Systems 2020 (MWSCAS '20). [Paper]
  4. Chandan Kumar Jha, Kailash Prasad, Arun Singh Tomar, and Joycee Mekie, "SEDAAF: FPGA Based Single Exact Dual Approximate Adders for Approximate Processors," Accepted in IEEE International Symposium on Circuits and Systems 2020 (ISCAS '20). [Paper]
  5. Chandan Kumar Jha*, Kailash Prasad*, Vibhor Kumar Srivastava, and Joycee Mekie, "FPAD: A Multistage Approximation Methodology for Designing Floating Point Approximate Dividers," Accepted in IEEE International Symposium on Circuits and Systems 2020 (ISCAS '20). [Paper]
  6. Chandan Kumar Jha, Ankita Nandi, and Joycee Mekie, "Quality Tunable Approximate Adder for Low Energy Image Processing Applications," Accepted in IEEE International Conference on Electronics Circuits and Systems 2019 (ICECS '19). [Paper]
  7. Ankita Nandi, Chandan Kumar Jha, and Joycee Mekie, "Should We Code Differently When Using Approximate Circuits?," 2019 Asia Pacific Conference on Circuits and Systems (APCCAS '19). [Paper]
  8. Chandan Kumar Jha and Joycee Mekie, "Design of Novel CMOS based Inexact Subtractors and Dividers for Approximate Computing: An In-Depth Comparison with PTL based Designs," 2019 22nd Euromicro Conference on Digital System Design (DSD '19). [Paper]
  9. Chandan Kumar Jha and Joycee Mekie, "SEDA - Single Exact Dual Approximate Adders for Approximate Processors," 2019 56th Annual Design Automation Conference (DAC '19). [Paper]

Journals

  1. Chandan Kumar Jha, Shreyas Singh, Riddhi Thakker, Manu Awasthi and Joycee Mekie, "Zero Aware Configurable Data Encoding by Skipping Transfer for Error Resilient Applications," in IEEE Transactions on Circuits and Systems I (2021) (TCAS-I).[Preprint]
  2. Chandan Kumar Jha, Ishita Doshi and Joycee Mekie, "Analysis of Worst-Case Data Dependent Temporal Approximation in Floating Point Units," in IEEE Transactions on Circuits and Systems II: Express Briefs (2020) (TCAS-II). [Paper]
  3. Chandan Kumar Jha, Sneha N. Ved, Ishant Anand and Joycee Mekie, "Energy and Error Analysis Framework for Approximate Computing in Mobile Applications," in IEEE Transactions on Circuits and Systems II: Express Briefs (2019) (TCAS-II). [Paper]

Posters

  1. Chandan Kumar Jha, Sumit Walia, Gagan Kanojia, and Joycee Mekie, "Floating Point Configurable Approximate Multiplier," Work-in-Progress, 56th Annual Design Automation Conference 2019 (DAC '19).
      2. * Authors have contributed equally