Page 389 - 《软件学报》2025年第10期

4786 软件学报 2025 年第 36 卷第 10 期

6 总 结

本文将计算机系统领域编译基础设施的新进展 MLIR 与类脑计算领域对于可扩展、可复用编译工具的需求
结合起来, 设计实现了一个类脑计算编译框架及其验证原型——BIVM. BIVM 充分利用 MLIR 的 progressivity 特
性, 所设计的中间表示能够混合不同级别的抽象层次和概念, 而在高层到低层抽象的渐进递降过程中可灵活组合
不同层次的编译优化方法, 从而较好地解决了编译框架所面临的类脑计算应用特征跨度大、芯片体系结构跨度大
等问题. 测试表明, BIVM 编译生成程序的运行性能比目前广泛采用的开发框架更高 (针对处理器芯片), 或者比其
原有工具链生成的程序更高或相近 (针对所支持的两种类脑芯片).

References:
[1] Goertzel B. Artificial general intelligence: Concept, state of the art, and future prospects. Journal of Artificial General Intelligence, 2014,
5(1): 1–48. [doi: 10.2478/jagi-2014-0001]
[2] Hennessy JL, Patterson DA. A new golden age for computer architecture. Communications of the ACM, 2019, 62(2): 48–60. [doi: 10.
1145/3282307]
[3] Roy K, Jaiswal A, Panda P. Towards spike-based machine intelligence with neuromorphic computing. Nature, 2019, 575(7784): 607–617.
[doi: 10.1038/s41586-019-1677-2]
[4] Qu P, Yan J, Zhang YH, Gao GR. Parallel turing machine, a proposal. Journal of Computer Science and Technology, 2017, 32(2):
269–285. [doi: 10.1007/s11390-017-1721-3]
[5] Rhodes O. Brain-inspired computing boosted by new concept of completeness. Nature, 2020, 586(7829): 364–366. [doi: 10.1038/d41586-
020-02829-w]
[6] Hazan H, Saunders DJ, Khan H, Patel D, Sanghavi DT, Siegelmann HT, Kozma R. BindsNET: A machine learning-oriented spiking
neural networks library in Python. Frontiers in Neuroinformatics, 2018, 12: 89. [doi: 10.3389/fninf.2018.00089]
[7] Pehle CG, Pedersen JE. Norse—A deep learning library for spiking neural networks. 2021. https://zenodo.org/records/4422025
[8] Mozafari M, Ganjtabesh M, Nowzari-Dalini A, Masquelier T. SpykeTorch: Efficient simulation of convolutional spiking neural networks
with at most one spike per neuron. Frontiers in Neuroscience, 2019, 13: 625. [doi: 10.3389/fnins.2019.00625]
[9] Fang W, Chen YQ, Ding JH, Yu ZF, Masquelier T, Chen D, Huang LW, Zhou HH, Li GQ, Tian YH. SpikingJelly: An open-source
machine learning infrastructure platform for spike-based intelligence. Science Advances, 2023, 9(40): eadi1480. [doi: 10.1126/sciadv.
adi1480]
[10] Zhang YH, Qu P, Ji Y, Zhang WH, Gao GR, Wang GR, Song S, Li GQ, Chen WG, Zheng WM, Chen F, Pei J, Zhao R, Zhao MG, Shi
LP. A system hierarchy for brain-inspired computing. Nature, 2020, 586(7829): 378–384. [doi: 10.1038/s41586-020-2782-y]
[11] Richmond P, Cope A, Gurney K, Allerton DJ. From model specification to simulation of biologically constrained networks of spiking
neurons. Neuroinformatics, 2014, 12(2): 307–323. [doi: 10.1007/s12021-013-9208-z]
[12] Aimone JB, Severa W, Vineyard CM. Composing neural algorithms with Fugu. In: Proc. of the 2019 Int’l Conf. on Neuromorphic
Systems. Knoxville: ACM, 2019. 3. [doi: 10.1145/3354265.3354268]
[13] Lagorce X, Benosman R. STICK: Spike time interval computational kernel, a framework for general purpose computation using neurons,
precise timing, delays, and synchrony. Neural Computation, 2015, 27(11): 2261–2317. [doi: 10.1162/NECO_a_00783]
[14] Chen JJ, Yang L, Zhang YH. GaBAN: A generic and flexibly programmable vector neuro-processor on FPGA. In: Proc. of the 59th
ACM/IEEE Design Automation Conf. San Francisco: ACM, 2022. 931–936. [doi: 10.1145/3489517.3530561]
[15] Fei X, Zhang YH, Zheng WM. XB-SIM*: A simulation framework for modeling and exploration of ReRAM-based CNN acceleration
design. Tsinghua Science and Technology, 2021, 26(3): 322–334. [doi: 10.26599/TST.2019.9010070]
[16] Brette R, Rudolph M, Carnevale T, et al. Simulation of networks of spiking neurons: A review of tools and strategies. Journal of
Computational Neuroscience, 2007, 23(3): 349–398. [doi: 10.1007/s10827-007-0038-6]
[17] Burkitt AN. A review of the integrate-and-fire neuron model: I. Homogeneous synaptic input. Biological Cybernetics, 2006, 95(1): 1–19.
[doi: 10.1007/s00422-006-0068-6]
[18] Izhikevich EM. Simple model of spiking neurons. IEEE Trans. on Neural Networks, 2003, 14(6): 1569–1572. [doi: 10.1109/TNN.2003.
820440]
[19] Hodgkin AL, Huxley AF. A quantitative description of membrane current and its application to conduction and excitation in nerve.
Bulletin of Mathematical Biology, 1990, 52(1–2): 25–71. [doi: 10.1016/S0092-8240(05)80004-7]
[20] Carnevale NT, Hines ML. The NEURON Book. Cambridge: Cambridge University Press, 2006.
[21] Olshausen BA, Field DJ. How close are we to understanding V1? Neural Computation, 2005, 17(8): 1665–1699. [doi: 10.1162/
0899766054026639]