[其他] ICML-Training Quantized Neural Networks to Global Optimality via Semidefini ... [推广有奖]

1关注
167
粉丝

已卖：5673份资源

泰斗

还不是VIP/贵宾

威望: 0 级
论坛币: 13312 个
通用积分: 711.3950
学术水平: 459 点
热心指数: 484 点
信用等级: 414 点
经验: 116189 点
帖子: 4684
精华: 0
在线时间: 16723 小时
注册时间: 2013-1-2
最后登录: 2026-1-8

楼主

Mujahida

发表于 2025-7-27 08:40:54 |AI写论文

是否 +2 论坛币

k人参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

Training Quantized Neural Networks to Global Optimality via Semidefinite
            Programming

                     Burak Bartan 1 Mert Pilanci 1

         Abstract             dimensions using semidefinite programming (Bartan & Pi-
Neural networks (NNs) have been extremely suc-    lanci, 2021). In this work, we take a similar convex duality
cessful across many tasks in machine learning.    approach that involves semidefinite programming. However,
Quantization of NN ...