SFFAI论坛已开放注册

详情点击查看：https://bbs.sffai.com/d/312

尽管深度学习模型在大量任务中拥有极佳的性能，但它容易学习到虚假相关性。为了缓解这个问题，本期讲者王心怡同学提出了一个基于因果关系的训练框架，以减少由观察到的混杂因素而引起的虚假相关性。

讲者介绍

王心怡，加州大学圣塔芭芭拉分校博士生，主要研究方向为机器学习和自然语言处理，使用因果推理工具分析解决机器学习中的问题。

分享题目

利用反事实最大似然估计训练深度学习网络

分享摘要

We propose a causality-based training framework to reduce the spurious correlations caused by observed confounders. We give theoretical analysis on the underlying general Structural Causal Model (SCM) and propose to perform Maximum Likelihood Estimation (MLE) on the interventional distribution instead of the observational distribution, namely Counterfactual Maximum Likelihood Estimation (CMLE). As the interventional distribution, in general, is hidden from the observational data, we then derive two different upper bounds of the expected negative log-likelihood and propose two general algorithms, Implicit CMLE and Explicit CMLE, for causal predictions of deep learning models using observational data. We conduct experiments on both simulated data and two real-world tasks: Natural Language Inference (NLI) and Image Captioning. The results show that CMLE methods outperform the regular MLE method in terms of out-of-domain generalization performance and reducing spurious correlations, while maintaining comparable performance on the regular evaluations.

论文题目：Counterfactual Maximum Likelihood Estimation for Training Deep Networks

分享亮点

1. We formalize the spurious correlation problem as a confounding problem using a structural causal model.

2. We propose a new general training scheme to reduce the unwanted effect of observed confounders with provable bounds.

3. The proposed method is evaluated on both simulated data and real-world tasks and is shown to be superior to the regular ERM training scheme in terms of out-of-domain generalization performance and reducing spurious correlations, while maintaining comparable performance on the regular evaluations.

直播时间

2022年2月27日（周日）21:00—22:00 线上直播

关注微信公众号：人工智能前沿讲习，对话框回复“SFFAI138”，获取入群二维码

人工智能前沿学生论坛 SFFAI（Student Forums on Frontiers of Artificial Intelligence）每周日举行一期， 邀请一线科研人员分享、讨论人工智能各个领域的前沿思想和最新成果，使专注于各个细分领域的研究者开拓视野、触类旁通。SFFAI 目前主要关注计算机视觉、自然语言处理、数据挖掘、机器学习理论等各个人工智能垂直领域及交叉领域的前沿进展，进行学术传播，为初学者指明方向，也为讲者塑造个人影响力。

欢迎大家关注SFFAI论坛：https://bbs.sffai.com/。