paper-with-me

홈 › Papers

The Double-Ellipsoid Geometry of CLIP

2024-11-21 · Meir Yossef Levi, Guy Gilboa

Contrastive Language-Image Pre-Training (CLIP) is highly instrumental in machine learning applications within a large variety of domains. We investigate the geometry of this embedding, which is still not well understood. We examine the raw unnormalized embedding and show that text and image reside on linearly separable ellipsoid shells, not centered at the origin. We explain the benefits of having this structure, allowing to better embed instances according to their uncertainty during contrastive training. Frequent concepts in the dataset yield more false negatives, inducing greater uncertainty. A new notion of conformity is introduced, which measures the average cosine similarity of an instance to any other instance within a representative data set. We show this measure can be accurately estimated by simply computing the cosine similarity to the modality mean vector. Furthermore, we find that CLIP's modality gap optimizes the matching of the conformity distributions of image and text.

📄 PDF Abstract BibTeX arXiv:2411.14517

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Compressive classification and the rare eclipse problem

2014-04-11 · Afonso S. Bandeira, Dustin G. Mixon, Benjamin Recht

This paper addresses the fundamental question of when convex sets remain disjoint after random projection. We provide an analysis using ideas from high-dimensional convex geometry. For ellipsoids, we provide a bound in t…

ClassificationGeneral Classification

EllipsoidNet: Ellipsoid Representation for Point Cloud Classification and Segmentation

2021-03-03 · Yecheng Lyu, Xinming Huang, Ziming Zhang

Point cloud patterns are hard to learn because of the implicit local geometry features among the orderless points. In recent years, point cloud representation in 2D space has attracted increasing research interest since …

ClassificationGeneral ClassificationPoint Cloud Classification

Compact 3D Gaussian Splatting For Dense Visual SLAM

2024-03-17 · Tianchen Deng, Yaohui Chen, Leyan Zhang, Jianfei Yang 외

Recent work has shown that 3D Gaussian-based SLAM enables high-quality reconstruction, accurate pose estimation, and real-time rendering of scenes. However, these approaches are built on a tremendous number of redundant …

Pose Estimation

Action Candidate Driven Clipped Double Q-learning for Discrete and Continuous Action Tasks

2022-03-22 · Haobo Jiang, Jin Xie, Jian Yang

Double Q-learning is a popular reinforcement learning algorithm in Markov decision process (MDP) problems. Clipped Double Q-learning, as an effective variant of Double Q-learning, employs the clipped double estimator to …

Q-Learning

Action Candidate Based Clipped Double Q-learning for Discrete and Continuous Action Tasks

2021-05-03 · Haobo Jiang, Jin Xie, Jian Yang

Double Q-learning is a popular reinforcement learning algorithm in Markov decision process (MDP) problems. Clipped Double Q-learning, as an effective variant of Double Q-learning, employs the clipped double estimator to …

Q-Learning