Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

INTELLIGENT DEVICES

Lego-like reconfigurable AI chips

Nature Electronics volume 5, pages 327–328 (2022)Cite this article

Subjects

A reconfigurable integration technology based on stackable chips with embedded arrays of optoelectronic devices and memristive crossbars could be of use in edge intelligence applications.

There is an increasing demand to be able to process sensory information in real time and at a low power. This, however, requires the development and integration of different types of intelligent device, from sensors to processors. Conventional three-dimensional (3D) heterogeneous integration technology, which is based on a von Neumann architecture, relies on the use of metal wire vias to vertically connect sensor, processor and memory layers1. In such systems, the hardwired connections between these distinct functional layers are fixed once fabricated. As a result, conventional 3D heterogeneous integration technology is inefficient when used in edge intelligence applications, where time-varying demands for sensing and computing are widespread. Writing in Nature Electronics, Huaqiang Wu, Peng Lin, Jeehwan Kim and colleagues now report a reconfigurable heterogeneous integration technology that is based on stackable chips with embedded arrays of optoelectronic devices and memristive crossbars2.

This is a preview of subscription content

Access options

Buy article

Get time limited or full article access on ReadCube.

$32.00

Buy

All prices are NET prices.

Fig. 1: Stackable hetero-integrated chips with light communication.

References

  1. Shulaker, M. et al. Nature 547, 74–78 (2017).

    Article  Google Scholar 

  2. Choi, C. et al. Nat. Electron. https://doi.org/10.1038/s41928-022-00778-y (2022).

    Article  Google Scholar 

  3. Zhou, F. et al. Nat. Nanotechnol. 14, 776–782 (2019).

    Article  Google Scholar 

  4. Mennel, L. et al. Nature 579, 62–66 (2020).

    Article  Google Scholar 

  5. Wang, C. et al. Sci. Adv. 6, eaba6173 (2020).

    Article  Google Scholar 

  6. Yao, P. et al. Nature 577, 641–646 (2020).

    Article  Google Scholar 

  7. Lin, P. et al. Nat. Electron. 3, 225–232 (2020).

    Article  Google Scholar 

  8. Wang, S. et al. Natl Sci. Rev. 8, nwaa172 (2021).

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Institute of Brain-Inspired Intelligence, National Laboratory of Solid State Microstructures, School of Physics, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing, China

    Shi-Jun Liang & Feng Miao

Authors
  1. Shi-Jun Liang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Feng Miao
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Feng Miao.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Liang, SJ., Miao, F. Lego-like reconfigurable AI chips. Nat Electron 5, 327–328 (2022). https://doi.org/10.1038/s41928-022-00785-z

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41928-022-00785-z

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing