https://hackerpaper.kaipfister.de/p/1607.06450/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=47785367urn:hackerpapers:mention:1607.06450:47785367Wed, 15 Apr 2026 21:18:47 GMT<p>Sure; https://arxiv.org/pdf/1607.06450 Depending on the model architecture, there is normalization taking place in multiple different places in order to save compute and ensure (some) consistency in output. Training, by its very nature, also is a normalization function, since you are telling the model which outputs are and are not valid, shaping weights that define features.</p><p><a href="https://hackerpaper.kaipfister.de/p/1607.06450/">Paper and discussion history</a>. A mention is not an endorsement.</p>stat.MLhttps://news.ycombinator.com/item?id=20865268urn:hackerpapers:mention:1607.06450:20865268Tue, 03 Sep 2019 10:44:07 GMT<p>Isn&#x27;t this outdated retro technology compared to layer normalization? &gt;Training state-of-the-art, deep neural networks is computationally expensive. One way to reduce the training time is to normalize the activities of the neurons. A recently introduced technique called batch normalization uses the distribution of the summed input to a neuron over a mini-batch of training cases to compute a mean and variance which are then used to normalize the summed input to that neuron on each training case. This significantly reduces the training time in feed-forward neural networks. However, the effect…</p><p><a href="https://hackerpaper.kaipfister.de/p/1607.06450/">Paper and discussion history</a>. A mention is not an endorsement.</p>stat.ML