文章预览
LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 1、[CV] What Makes a Maze Look Like a Maze? 2、[LG] MarS:a Financial Market Simulation Engine Powered by Generative Foundation Model 3、[LG] Heterogeneous Sheaf Neural Networks 4、[CL] Source2Synth:Synthetic Data Generation and Curation Grounded in Real Data Sources 5、[LG] Ratio Divergence Learning Using Target Energy in Restricted Boltzmann Machines:Beyond Kullback--Leibler Divergence Learning 摘要:是什么让迷宫看起来像迷宫、由生成式基础模型支持的金融市场模拟引擎、异构片状神经网络、基于真实数据源的合成数据生成和整理、在受限玻尔兹曼机中使用目标能量进行比率散度学习 1、[CV] What Makes a Maze Look Like a Maze? J Hsu, J Mao, J B. Tenenbaum, N D. Goodman… [Stanford University & MIT] 是什么让迷宫看起来像迷宫? 要点: 提出一个在图像中理解抽象概念的视觉推理任务,它需要抽象的规则而不是具
………………………………