no code implementations • 25 May 2024 • Chenqi Lin, Tianshi Xu, Zebin Yang, Runsheng Wang, Ru Huang, Meng Li
We observe the overhead mainly comes from the neglect of 1) the one-hot nature of user queries and 2) the robustness of the embedding table to low bit-width quantization noise.
no code implementations • 21 Feb 2024 • Shuzhang Zhong, Zebin Yang, Meng Li, Ruihao Gong, Runsheng Wang, Ru Huang
Additionally, it introduces a dynamic token tree generation algorithm to balance the computation and parallelism of the verification phase in real-time and maximize the overall efficiency across different batch sizes, sequence lengths, and tasks, etc.
no code implementations • 21 Jan 2024 • Rongqing Cong, Wenyang He, Mingxuan Li, Bangning Luo, Zebin Yang, Yuchao Yang, Ru Huang, Bonan Yan
Large language models (LLMs) with Transformer architectures have become phenomenal in natural language processing, multimodal generative artificial intelligence, and agent-oriented artificial intelligence.
1 code implementation • 7 May 2023 • Agus Sudjianto, Aijun Zhang, Zebin Yang, Yu Su, Ningzhou Zeng
PiML (read $\pi$-ML, /`pai`em`el/) is an integrated and open-access Python toolbox for interpretable machine learning model development and model diagnostics.
no code implementations • 15 Dec 2020 • Yifeng Guo, Yu Su, Zebin Yang, Aijun Zhang
In this paper, we propose the explainable recommendation systems based on a generalized additive model with manifest and latent interactions (GAMMLI).
1 code implementation • 8 Nov 2020 • Agus Sudjianto, William Knauth, Rahul Singh, Zebin Yang, Aijun Zhang
We propose the local linear profile plot and other visualization methods for interpretation and diagnostics, and an effective merging strategy for network simplification.
2 code implementations • 8 Sep 2020 • Zebin Yang, Aijun Zhang
Hyperparameter optimization (HPO) plays a central role in the automated machine learning (AutoML).
no code implementations • 16 May 2020 • Zebin Yang, Hengtao Zhang, Agus Sudjianto, Aijun Zhang
Network initialization is the first and critical step for training neural networks.
2 code implementations • 16 Mar 2020 • Zebin Yang, Aijun Zhang, Agus Sudjianto
The lack of interpretability is an inevitable problem when using neural network models in real applications.
no code implementations • 12 Jan 2019 • Zebin Yang, Aijun Zhang, Agus Sudjianto
It leads to an explainable neural network (xNN) with the superior balance between prediction performance and model interpretability.