https://3rdlayer.uk/tags/deep-learning/Recent content in Deep-Learning on 3rd layerHugo -- 0.154.5en-USFri, 03 Jul 2026 00:00:00 +0000https://3rdlayer.uk/posts/relu-piecewise-affine/Fri, 03 Jul 2026 00:00:00 +0000https://3rdlayer.uk/posts/relu-piecewise-affine/A network with ReLU activations secretly carves input space into polytope pieces and, on each piece, collapses into exactly one affine map. This post builds that view up from a single neuron to interactive demos.https://3rdlayer.uk/posts/kmeans-weight-quantization/Sat, 06 Jun 2026 00:00:00 +0000https://3rdlayer.uk/posts/kmeans-weight-quantization/Group a neural network’s weights into a few representative values with K-Means, and you can shrink the model several-fold with almost no accuracy loss. We explore it with an interactive widget where you watch the clusters converge and the storage shrink in real time. (Deep Compression, Han et al. 2016)https://3rdlayer.uk/posts/linear-quantization/Sat, 09 May 2026 00:00:00 +0000https://3rdlayer.uk/posts/linear-quantization/Beyond storing weights as integers — running the multiplications and additions entirely in integer arithmetic at inference. We connect reals and integers with the affine map r = S(q − Z), and explore it with a widget where you change the scale and zero point and watch the quantization error. (Jacob et al. 2018, the basis of TFLite integer quantization)https://3rdlayer.uk/posts/numeric-data-types/Sun, 12 Apr 2026 00:00:00 +0000https://3rdlayer.uk/posts/numeric-data-types/INT8, FP16, BF16, FP8, FP4 — what do the data types you keep seeing in deep learning actually mean, and how do bits turn into numbers? We take them apart one by one, with widgets where clicking a bit updates the formula and value in real time.