A variation-tolerant in-memory machine learning classifier via on-chip training

Sujan K. Gonugondla, Mingu Kang, Naresh R. Shanbhag

Research output: Contribution to journalArticlepeer-review


This paper presents a robust deep in-memory machine learning classifier with a stochastic gradient descent (SGD)-based on-chip trainer using a standard 16-kB 6T SRAM array. The deep in-memory architecture (DIMA) enhances both energy efficiency and throughput over conventional digital architectures by reading multiple bits per bit line (BL) per read cycle and by employing mixed-signal processing in the periphery of the bit-cell array. Though these techniques improve the energy efficiency and latency, DIMA's analog nature makes it sensitive to process, voltage, and temperature (PVT) variations, especially under reduced BL swings. On-chip training enables DIMA to adapt to chip-specific variations in PVT as well as data statistics, thereby further enhancing its energy efficiency. The 65-nm CMOS prototype IC demonstrates this improvement by realizing an on-chip trainable support vector machine. By learning chip-specific weights, on-chip training enables robust operation under reduced BL swing leading to a 2.4 times reduction in energy over an off-chip trained DIMA. The prototype IC in 65-nm CMOS consumes 42 pJ/decision at 32 M decisions/s, corresponding to 3.12 TOPS/W (1 OP = one 8-b × 8-b MAC) during inference, thereby achieving a reduction of 21 times in energy and 100 times in energy-delay product as compared with a conventional digital architecture. The energy overhead of training is <26% per decision for SGD batch sizes of 128 and higher.

Original languageEnglish (US)
Article number8463601
Pages (from-to)3163-3173
Number of pages11
JournalIEEE Journal of Solid-State Circuits
Issue number11
StatePublished - Nov 2018
Externally publishedYes


  • Always-on
  • in-memory
  • inference
  • machine learning (ML)
  • mixed signal
  • on-chip learning
  • process variations
  • stochastic gradient descent (SGD)

ASJC Scopus subject areas

  • Electrical and Electronic Engineering


Dive into the research topics of 'A variation-tolerant in-memory machine learning classifier via on-chip training'. Together they form a unique fingerprint.

Cite this