Want to create an interactive transcript for this episode?
Podcast: Arxiv Papers
Episode: [QA] Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training
Description: This paper compares cosine annealing and infinite learning rate schedules for continual self-supervised learning, finding the latter more effective in enhancing pre-training performance across various datasets.https://arxiv.org/abs//2503.02844YouTube: https://www.youtube.com/@ArxivPapersTikTok: https://www.tiktok.com/@arxiv_papersApple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers