{"total":47,"papers":[{"id":"url:iclr.cc/virtual/2025/poster/30565","title":"ADAM Optimization with Adaptive Batch Selection","date":"2024-10-01","section":"Data, training, optimization","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/484127f73a5957a2fbcab07a704fa688492d75165463a4d52f34ba421316f7f9/","source_url":"https://iclr.cc/virtual/2025/poster/30565","text_available":false},{"id":"arxiv:2312.10549","title":"Catastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy","date":"2023-12-16","section":"Data, training, optimization","quality":-0.651376,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/890b8ca392bd314b377271caffd70091a482d479f8cc7747c31633479b2e87d8/","source_url":"https://arxiv.org/abs/2312.10549","text_available":true},{"id":"arxiv:2108.06325","title":"Continual Backprop: Stochastic Gradient Descent with Persistent Randomness","date":"2021-08-13","section":"Data, training, optimization","quality":-0.471382,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/73c90c52bffd8b5b2485d93607350545d822f361c29d3c0063b8d3025576e793/","source_url":"https://arxiv.org/abs/2108.06325","text_available":true},{"id":"arxiv:2403.05175","title":"Continual Learning and Catastrophic Forgetting","date":"2024-03-08","section":"Data, training, optimization","quality":-0.26995,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3ed9ae586e8898d0235553d62c5f002609d3d112cbec54cc46655d19ff736400/","source_url":"https://arxiv.org/abs/2403.05175","text_available":true},{"id":"arxiv:2212.14034","title":"Cramming: Training a Language Model on a Single GPU in One Day","date":"2022-12-28","section":"Data, training, optimization","quality":-0.087484,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/cb243af26cd29bcc3d04b00cc68393b3c4b3dfc535c3dd228b44f12e5df26c46/","source_url":"https://arxiv.org/abs/2212.14034","text_available":true},{"id":"arxiv:1506.01186","title":"Cyclical Learning Rates for Training Neural Networks","date":"2015-06-03","section":"Data, training, optimization","quality":-0.406646,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/e05cf550191656c287184521d0ff011092f5ed6baa72e50f3a5bfd25d0f5986a/","source_url":"https://arxiv.org/abs/1506.01186","text_available":true},{"id":"url:openai.com/index/deep-double-descent","title":"Deep Double Descent (OpenAI)","date":"2020-01-23","section":"Data, training, optimization","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/dbfa3f03c433f18313852b6668892ef9cbe993f6949854e9270622c7cd29b701/","source_url":"https://openai.com/index/deep-double-descent/","text_available":false},{"id":"arxiv:2410.07041","title":"Emergent properties with repeated examples","date":"2024-10-09","section":"Data, training, optimization","quality":0.087466,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/63a5d833b43fae5ed1f9f44f26b90813fa8a37dd44870e8e409d7a7c34fabfec/","source_url":"https://arxiv.org/abs/2410.07041","text_available":true},{"id":"arxiv:2607.27372","title":"Explorative Modeling: Unlocking a Third Pretraining Axis and End-to-End Generation","date":"2026-07-29","section":"Data, training, optimization","quality":0.296593,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/542a5e85b84624381953f643ec6075a588c08c542c4f945e57d39c9164c9a835/","source_url":"https://arxiv.org/abs/2607.27372","text_available":true},{"id":"arxiv:2405.20233","title":"Grokfast: Accelerated Grokking by Amplifying Slow Gradients","date":"2024-05-30","section":"Data, training, optimization","quality":0.183974,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/b62d50c44fbd3554702d96f717a5ee25029a28ceed9346ab27ff64d1d08ba07e/","source_url":"https://arxiv.org/abs/2405.20233","text_available":true},{"id":"arxiv:2405.16684","title":"gzip Predicts Data-dependent Scaling Laws","date":"2024-05-26","section":"Data, training, optimization","quality":-0.037016,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/8d5456f36b61f2ce9a45da0889b5f0318dc816cab90ee57d54bba06246041de3/","source_url":"https://arxiv.org/abs/2405.16684","text_available":true},{"id":"arxiv:2505.24832","title":"How much do language models memorize?","date":"2025-05-30","section":"Data, training, optimization","quality":0.266715,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/1d14faeb604373b83ad90ae653998a11cc09492020341f29419e0deed45a300c/","source_url":"https://arxiv.org/abs/2505.24832","text_available":true},{"id":"url:educatingsilicon.com/2024/05/09/how-much-llm-training-data-is-there-in-the-limit","title":"How much LLM training data is there, in the limit?","date":"2024-05-09","section":"Data, training, optimization","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/1b8290cc5831de1490685a90105ebb81f55a67ebb3178e4fabac47caac2bcb43/","source_url":"https://www.educatingsilicon.com/2024/05/09/how-much-llm-training-data-is-there-in-the-limit/","text_available":false},{"id":"arxiv:2009.11848","title":"How Neural Networks Extrapolate: From Feedforward to Graph Neural Networks","date":"2020-09-24","section":"Data, training, optimization","quality":0.221265,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/bd1a0149b2db51763f5d8c949f315e618ad47d5f1e2976b08a049f2bed2fbc4c/","source_url":"https://arxiv.org/abs/2009.11848","text_available":true},{"id":"arxiv:1904.00962","title":"Large Batch Optimization for Deep Learning: Training BERT in 76 minutes","date":"2019-04-01","section":"Data, training, optimization","quality":0.469974,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/2cea31116143e748e7497985367ddcf86a3da4aadb24ea47451b55947eedc9f0/","source_url":"https://arxiv.org/abs/1904.00962","text_available":true},{"id":"arxiv:2110.09485","title":"Learning in High Dimension Always Amounts to Extrapolation","date":"2021-10-18","section":"Data, training, optimization","quality":-0.296541,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/589d10ae90bda37ab0288fb5eadf87b368c249900944a0fe8b21a2f9bb267c01/","source_url":"https://arxiv.org/abs/2110.09485","text_available":true},{"id":"arxiv:2312.17742","title":"Learning Vision from Models Rivals Learning Vision from Data","date":"2023-12-28","section":"Data, training, optimization","quality":0.170732,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/","source_url":"https://arxiv.org/abs/2312.17742","text_available":true},{"id":"doi:10.1038/s41586-024-07711-7","title":"Maintaining Plasticity in Deep Continual Learning","date":"2023-06-23","section":"Data, training, optimization","quality":0.362714,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/b5d46327415fe38f2b9160b874735530939961a71536f7f1ac4c0dc7a3af3224/","source_url":"https://doi.org/10.1038/s41586-024-07711-7","text_available":true},{"id":"arxiv:1708.02072","title":"Measuring Catastrophic Forgetting in Neural Networks","date":"2017-08-07","section":"Data, training, optimization","quality":-0.495708,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/358d4d451093ba5234c459f6323da4968d71d8be905b6612f5bae0162e9145f2/","source_url":"https://arxiv.org/abs/1708.02072","text_available":true},{"id":"arxiv:2402.02342","title":"MetaOptimize: A Framework for Optimizing Step Sizes and Other Meta-parameters","date":"2024-02-04","section":"Data, training, optimization","quality":-0.208417,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/6801d4fe20b85924aed2dec1855a35232b2ec1d98e15ec3bc27a0fbe5dfd1af7/","source_url":"https://arxiv.org/abs/2402.02342","text_available":true},{"id":"arxiv:2512.24695","title":"Nested Learning: The Illusion of Deep Learning Architectures","date":"2025-12-31","section":"Data, training, optimization","quality":-0.629827,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d19bf2367894704a701d0741e7db6d8b148719af1d634b55f5797aab5f617d06/","source_url":"https://arxiv.org/abs/2512.24695","text_available":true},{"id":"arxiv:2307.06440","title":"No Train No Gain: Revisiting Efficient Training Algorithms For Transformer-based Language Models","date":"2023-07-12","section":"Data, training, optimization","quality":0.017146,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/fc9b59e6276d3fc1ebf8d77b30466c5047df876d0a884edfdfcdea124d6b76d6/","source_url":"https://arxiv.org/abs/2307.06440","text_available":true},{"id":"arxiv:2510.05491","title":"NorMuon: Making Muon more efficient and scalable","date":"2025-10-07","section":"Data, training, optimization","quality":0.070672,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/c9ea441f9c2808077269690a88c1b1407189ac1cb03fbc0df89979a4d1d41b98/","source_url":"https://arxiv.org/abs/2510.05491","text_available":true},{"id":"openreview:ry_WPG-A-","title":"On the Information Bottleneck Theory of Deep Learning (Saxe et al.)","date":"2017-10-27","section":"Data, training, optimization","quality":-0.011319,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/c10860ec07fd50dd9cabe10b023919ca199ae9bb2897cadbb523bbea39394ecf/","source_url":"https://openreview.net/forum?id=ry_WPG-A-","text_available":true}],"next":"https://vinnik-dmitry07.github.io/favorite-papers/agents/catalog-7-2.html"}