研究

我主要研究生成模型加速、神经形态计算与高效模型。研究兴趣涵盖脑启发计算(脉冲神经网络与神经形态系统)、高效建模(加速算法、剪枝与量化),以及 AI 安全(对抗攻击与模型可解释性)。

博士期间主要围绕脉冲神经网络、神经形态计算与高效模型开展研究。同时,我也在阿里巴巴 Wan 团队实习,探索视频生成模型的加速方法。

Google Scholar

已发表工作

2026

  1. Proxy target: Bridging the gap between discrete spiking neural networks and continuous control

    Z Xu, T Bu, Z Hao, J Ding, Z Yu

    Advances in Neural Information Processing Systems 38, 159158-159184会议论文

  2. Activity Pruning for Efficient Spiking Neural Networks

    T Bu, X Shi, Z Yu

    Advances in Neural Information Processing Systems 38, 171134-171153会议论文

  3. Generalized Threshold Optimization with Harmony Multi-Threshold Neurons for Accurate ANN-to-SNN Conversion

    W Zhang, Z Huang, T Bu, T Huang, Z Yu

    Proceedings of the AAAI Conference on Artificial Intelligence 40 (3), 2227-2235会议论文

  4. UniSparse: Combining Weight Pruning and Spike Sparsification in Spiking Neural Networks

    X Shi, T Bu, Z Yu

    The 43rd International Conference on Machine Learning会议论文

2025

  1. Ultrafast probabilistic neuron in an artificial spin ice for robust deep neural networks

    Z Liang, T Bu, Z Lyu, Z Liu, A Hrabec, L Wang, Y Dou, J Ding, P Ge, ...

    Advanced Functional Materials 35 (11), 2417334期刊论文

2024

  1. Towards high-performance spiking transformers from ANN to SNN conversion

    Z Huang, X Shi, Z Hao, T Bu, J Ding, Z Yu, T Huang

    Proceedings of the 32nd ACM International Conference on Multimedia, 10688-10697会议论文

  2. A progressive training framework for spiking neural networks with learnable multi-hierarchical model

    Z Hao, X Shi, Z Huang, T Bu, Z Yu, T Huang

    International Conference on Learning Representations 2024, 4925-4941会议论文

  3. Threaten spiking neural networks through combining rate and temporal information

    Z Hao, T Bu, X Shi, Z Huang, Z Yu, T Huang

    International Conference on Learning Representations 2024, 55658-55674会议论文

  4. Robust decoding of rich dynamical visual scenes with retinal spikes

    Z Yu, T Bu, Y Zhang, S Jia, T Huang, JK Liu

    IEEE Transactions on Neural Networks and Learning Systems 36 (2), 3396-3409期刊论文

2023

  1. Reducing ANN-SNN conversion error through residual membrane potential

    Z Hao, T Bu, J Ding, T Huang, Z Yu

    Proceedings of the AAAI Conference on Artificial Intelligence 37 (1), 11-21会议论文

  2. Rate gradient approximation attack threats deep spiking neural networks

    T Bu, J Ding, Z Hao, Z Yu

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition会议论文

  3. Red teaming deep neural networks with feature synthesis tools

    S Casper, T Bu, Y Li, J Li, K Zhang, K Hariharan, D Hadfield-Menell

    Advances in Neural Information Processing Systems 36, 80470-80516会议论文

2022

  1. Optimized potential initialization for low-latency spiking neural networks

    T Bu, J Ding, Z Yu, T Huang

    Proceedings of the AAAI Conference on Artificial Intelligence 36 (1), 11-20会议论文

  2. SNN-RAT: Robustness-enhanced spiking neural network through regularized adversarial training

    J Ding, T Bu, Z Yu, T Huang, J Liu

    Advances in Neural Information Processing Systems 35, 24780-24793会议论文

  3. Decoding pixel-level image features from two-photon calcium signals of macaque visual cortex

    Y Zhang, T Bu, J Zhang, S Tang, Z Yu, JK Liu, T Huang

    Neural Computation 34 (6), 1369-1397期刊论文