莫雷拉
原始洛拉:
$ W = W_0 + UV $和 $等级(UV) leq r $
更好的初始化:
$ W = W_0 -U_0 {V_0} + UV $
添加剂洛拉:
美元在哪里 $ u in mathbb {r}^{m times r},v in { mathbb {r}^{r times n}} $和 $等级(UV) LEQ 2R $
Hadamard Mul Lora:
$ w = w_0 + odot_ {i = 1}^{i = k}( delta_i)$在哪里 $ delta_i = u_iv_i $
$ r'= frac {r} {k},u_i in mathbb {r}^{m times r'} $ ,,,, $ v_i in mathbb {r}^{r' times n} $和 $等级( odot_ {i = 1}^{i = k}( delta_i^t)) leq( frac {r} {k} {k})^k $
Hadamard添加Lora:
$ w = w_0 + odot_ {i = 1}^{i = k}( delta_i)$在哪里美元
$ r'= frac {r} {k} $ ,,,, $ u_i in mathbb {r}^{r' times n} $ ,,,, $ v_i in mathbb {r}^{m times r'} $和 $等级( odot_ {i = 1}^{i = k}( delta_i)) leq( frac {2r} {k} {k})^k $
Hadamard Lora:激活
$ delta = odot_ {i = 1}^{i = k}( tanh(u_iv_i^t))$
$ delta = odot_ {i = 1}^{i = k}( sigma(u_iv_i^t))$
Dylora:
随机更新一系列排名
更新部分的部分
参考:
@online { kexuefm-9590 ,
title = {梯度视角下的LoRA:简介、分析、猜测及推广} ,
author = {苏剑林} ,
year = { 2023 } ,
month = { Apr } ,
url = { url{https://spaces.ac.cn/archives/9590} } ,
} @misc { hyeonwoo2023fedpara ,
title = { FedPara: Low-Rank Hadamard Product for Communication-Efficient Federated Learning } ,
author = { Nam Hyeon-Woo and Moon Ye-Bin and Tae-Hyun Oh } ,
year = { 2023 } ,
eprint = { 2108.06098 } ,
archivePrefix = { arXiv } ,
primaryClass = { cs.LG }
}