人工智能2018 · 分享于 Hacker News
The Lottery Ticket Hypothesis: Finding Small, Trainable Neural Networks
探索《The Lottery Ticket Hypothesis: Finding Small, Trainable Neural Networks》的原文与 Hacker News 社区讨论，查看来源并加入阅读清单。
从论文，走向讨论
看看社区如何讨论这个想法。
来自 Hacker News 的原始讨论。不同的视角、疑问，以及值得继续思考的线索。
最多展示 100 条相关讨论。论文可能出现在评论中，并非原帖投稿内容。
在讨论中发现的引用
按 HN 条目顺序展示最近的 50 条引用来源。
评论 #47884357 评论 #46057470 评论 #46016820 评论 #44945353 评论 #42957002 评论 #42876001 评论 #41843682 评论 #38476008 评论 #36189137 评论 #36143990 评论 #35808092 评论 #35113535 评论 #34572784 评论 #34572755 评论 #34561783 评论 #33347962 评论 #30289756 评论 #30078770 评论 #28649266 评论 #28608317 评论 #27967174 评论 #25489287 评论 #25355683 评论 #24372880 评论 #24372189 评论 #23685336 评论 #23119601 评论 #22650905 评论 #22512774 评论 #20828730 评论 #20076155 评论 #20030462 评论 #19966474 评论 #19935625 评论 #19847068
Ytkellogg ·
S1: A $6 R1 competitor?
851 赞同416 评论
Ysakopov ·
Ask HN: What projects are you working on now?
809 赞同1,236 评论
Ytimsuchanek ·
OpenAI says it has evidence DeepSeek used its model to train competitor
747 赞同1,541 评论
Yluizfelberti ·
Ask HN: What's the best paper you've read in 2020?
705 赞同192 评论
Ybirriel ·
Mammals dream about the world they are about to experience before they are born
578 赞同183 评论
Ytheafh ·
Computer scientists prove why bigger neural networks do better
282 赞同145 评论
Ydataminer ·
Zamba2-7B
282 赞同71 评论
Ywebmaven ·
Researchers propose faster, more efficient alternatives to backpropagation
239 赞同103 评论
Ylifeisstillgood ·
Statistical Process Control in Python
230 赞同73 评论
Yheadalgorithm ·
Researchers unveil a pruning algorithm to shrink deep learning models
208 赞同59 评论
Ymagoghm ·
Smarter Training of Neural Networks
204 赞同40 评论
Ylamchob ·
Exploring Weight Agnostic Neural Networks
179 赞同64 评论
Yche_shr_cat ·
Proving the Lottery Ticket Hypothesis: Pruning is All You Need
174 赞同61 评论
Ysebg ·
Learning@home hivemind – train large neural networks across the internet
148 赞同38 评论
YRageoftheRobots ·
Deep Learning's Diminishing Returns
142 赞同84 评论
Yjonbaer ·
The Lottery Ticket Hypothesis: Finding Small, Trainable Neural Networks
137 赞同14 评论
Y076ae80a-3c97-4 ·
The lottery ticket hypothesis: why neural networks work
135 赞同64 评论
Ypizza ·
Neural nets typically contain smaller “subnetworks” that can often learn faster
128 赞同20 评论
YAnon84 ·
Neurons that fire together, wire together, but how?
127 赞同45 评论
Ysebg ·
An overview of the theory of overparameterized machine learning
123 赞同52 评论
Yaxiom92 ·
Notes on Weight Initialization for Deep Neural Networks
117 赞同18 评论
Yminimaxir ·
The AI Research SuperCluster
92 赞同50 评论
YTomte ·
No Brainer (2015)
85 赞同14 评论
Y0x79de ·
Weight-sparse transformers have interpretable circuits [pdf]
78 赞同46 评论
Yhuijzer ·
Substitute certain overhyped programming terms with more accurate replacements
60 赞同43 评论
Ytosh ·
SparseGPT: Language Models Can Be Accurately Pruned in One-Shot
24 赞同1 评论
Ymkmk ·
See the First Complete Map of an Insect’s Brain
18 赞同1 评论
Ysimonw ·
It’s infuriatingly hard to understand how closed models train on their input
7 赞同1 评论
Yheadalgorithm ·
AI intensifies fight against ‘paper mills’ that churn out fake research
4 赞同0 评论
Yjonbaer ·
The Lottery Ticket Hypothesis: Training Pruned Neural Networks
2 赞同0 评论
Ysillysaurusx ·
I won an ML lottery ticket, then squandered it
2 赞同1 评论
Ywslh ·
Can Generalist Foundation Models Beat Special-Purpose Tuning?Medicine Case Study
2 赞同0 评论
Yplat12 ·
The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks (2019)
1 赞同0 评论
Yalexmolas ·
The Transformer Family
1 赞同0 评论
Ykristianp ·
I won an ML lottery ticket, then squandered it
1 赞同1 评论
Yfelineflock ·
The Lottery Ticket Hypothesis: finding sparse trainable NNs with 90% less params
1 赞同0 评论
YDARSHANFOFADIYA ·
Show HN: SparseLab–real sparse training(CSR+custom kernel) in PyTorch, CPU-first
1 赞同1 评论