reinforcement perturbations observations adversarial

[Pix2Pix] Image-to-Image Translation with Conditional Adversarial NetWorks

paper:https://arxiv.org/pdf/1611.07004.pdf [CVPR 2017] code: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix https://phillipi.github.io/pix2pi ......

《Generative Adversarial Nets》论文精读

#论文精读《Generative Adversarial Nets》 导言:生成模型是目前爆火的一个研究方向,据Microsoft对于ChatGPT-4的研究称“ChatGPT-4可以看成是通用型人工智能(AGI)的早期版本;其独特的推理能力和理解语义能力迅速在全球掀起了大模型研究的一股热潮。不仅仅 ......
Adversarial Generative 论文 Nets

Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations

郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! NeurIPS 2020 ......

论文解读《Interpolated Adversarial Training: Achieving robust neural networks without sacrificing too much accuracy》

论文信息 论文标题:Interpolated Adversarial Training: Achieving robust neural networks without sacrificing too much accuracy论文作者:Alex LambVikas VermaKenji Kawa ......

Teachable Reinforcement Learning via Advice Distillation

**发表时间:**2021 (NeurIPS 2021) **文章要点:**这篇文章提出了一种学习policy的监督范式,大概思路就是先结构化advice,然后先学习解释advice,再从advice中学policy。这个advice来自于外部的teacher,相当于一种human-in-the-l ......

CellOracle | in silico gene perturbation | 新旧世代的交替

目的:对我们的单细胞多组学数据作此分析,看那个de-diff的TF的敲除能够逆转分化方向。 科研永远是追新者的天堂,不解释。 tutorial:Tutorial - Read the Docs https://github.com/morris-lab/CellOracle Dissecting c ......
新旧 世代 perturbation CellOracle silico

Cycle GAN:Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

paper:https://arxiv.org/pdf/1703.10593.pdf [2017] code 参考: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix https://zhuanlan.zhihu.com/p/792211 ......

Adversarial Robust Deep Reinforcement Learning Requires Redefining Robustness

郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! ......

论文解读(VAT)《Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning》

论文信息 论文标题:Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning论文作者:Takeru Miyato, S. Maeda, Masanori Koya ......

论文解读(PGD)《Towards deep learning models resistant to adversarial attacks》

论文信息 论文标题:Towards deep learning models resistant to adversarial attacks论文作者:Aleksander Madry, Aleksandar Makelov, Ludwig Schmidt, Dimitris Tsipras, Ad ......

论文解读(FGSM)《Explaining and Harnessing Adversarial Examples》

论文信息 论文标题:Explaining and Harnessing Adversarial Examples论文作者:Ian J. Goodfellow, Jonathon Shlens, Christian Szegedy论文来源:ICLR 2015论文地址:download 论文代码:dow ......

迁移学习(PAT)《Pairwise Adversarial Training for Unsupervised Class-imbalanced Domain Adaptation》

论文信息 论文标题:Pairwise Adversarial Training for Unsupervised Class-imbalanced Domain Adaptation论文作者:Weili Shi, Ronghang Zhu, Sheng Li论文来源:KDD 2022论文地址:dow ......

论文解读( FGSM)《Adversarial training methods for semi-supervised text classification》

论文信息 论文标题:Adversarial training methods for semi-supervised text classification论文作者:Taekyung Kim论文来源:ICLR 2017论文地址:download 论文代码:download视屏讲解:click 1 背 ......

强化学习 Reinforcement Learning

强化学习 Reinforcement Learning 强化学习是一种机器学习思想,其关心一个智能体如何采取行动以达到最大化激励回报。 基本的强化学习模型以马尔可夫决策过程建模。 马尔可夫决策过程 Markov Decision Process 系统要素 A 行动空间; S状态空间; $P^a_{s ......
Reinforcement Learning

迁移学习()《Attract, Perturb, and Explore: Learning a Feature Alignment Network for Semi-supervised Domain Adaptation》

论文信息 论文标题:Attract, Perturb, and Explore: Learning a Feature Alignment Network for Semi-supervised Domain Adaptation论文作者:Taekyung Kim论文来源:2020 ECCV论文地址 ......

Demonstration-Conditioned Reinforcement Learning for Few-Shot Imitation

**发表时间:**2021(ICML 2021) **文章要点:**这篇文章提出了demonstration-conditioned reinforcement learning (DCRL)来做Few-Shot Imitation,将demonstration和当前状态作为输入,通过强化学习最大化 ......

迁移学习(CDA)《CDA:Contrastive-adversarial Domain Adaptation 》

论文信息 论文标题:CDA:Contrastive-adversarial Domain Adaptation论文作者:Nishant Yadav, M. Alam, Ahmed K. Farahat, Dipanjan Ghosh, Chetan Gupta, A. Ganguly论文来源:202 ......

观察者模式(Observer)

#1,观察者模式 观察者模式是一种行为设计模式, 允许你定义一种订阅机制, 可在对象事件发生时通知多个 “观察” 该对象的其他对象。 又被称为发布-订阅(Publish/Subscribe)模式,它定义了一种一对多的依赖关系,让多个观察者对象同时监听某一个主题对象。这个主题对象在状态变化时,会通知所 ......
观察者 Observer 模式

CH582 CH592 CH573 CH579 Observer增加白名单

白名单机制 白名单(white list)是BLE协议中最简单、直白的一种安全机制。其原理很简单,总结如下(前面的分析文章中都有介绍): 所谓的白名单,就是一组蓝牙地址; 通过白名单,可以只允许特定的蓝牙设备(白名单中列出的)扫描(Scan)、连接(connect)我们,也可以只扫描、连接特定的蓝牙 ......
CH Observer 名单 582 573

02.Deep Reinforcement Learning for Quantitative Trading Challenges and Opportunities

Deep Reinforcement Learning for Quantitative Trading Challenges and Opportunities 量化交易的深度强化学习:挑战与机遇 IEEE 背景 量化交易:量化交易是指借助现代统计学和数学的方法,利用计算机技术来进行交易的证券投资 ......

迁移学习(ADDA)《Adversarial Discriminative Domain Adaptation》【已复现迁移】

论文信息 论文标题:Adversarial Discriminative Domain Adaptation论文作者:Eric Tzeng, Judy Hoffman, Kate Saenko, Trevor Darrell论文来源:CVPR 2017论文地址:download 论文代码:downl ......

虚假新闻检测-迁移学习(CADM)《Unsupervised Domain Adaptation for COVID-19 Information Service with Contrastive Adversarial Domain Mixup》

论文信息 论文标题:Unsupervised Domain Adaptation for COVID-19 Information Service with Contrastive Adversarial Domain Mixup论文作者:Huimin Zeng, Zhenrui Yue, Ziyi ......

【机器学习】李宏毅——Adversarial Attack(对抗攻击)

本文主要介绍了Adversarial Attack(对抗攻击)当前的研究现状,包括如何攻击、攻击的类别,以及原始模型如何进行防御等相关知识点。 ......
Adversarial 机器 Attack
共113篇  :4/4页 首页上一页4下一页尾页