From 8b458071bd26c034dbd1add7fa3667af7899ecb5 Mon Sep 17 00:00:00 2001 From: Yuchen Pei Date: Thu, 24 Jun 2021 17:58:50 +1000 Subject: Moved things in assets/resources out to assets/ --- assets/1423graph.jpg | Bin 0 -> 50282 bytes assets/dpmm.png | Bin 0 -> 41295 bytes assets/fully-bayesian-mm.png | Bin 0 -> 37129 bytes assets/hmm.png | Bin 0 -> 57074 bytes assets/juggling.png | Bin 0 -> 4802462 bytes assets/lda.png | Bin 0 -> 43373 bytes assets/mixture-model.png | Bin 0 -> 18380 bytes assets/plsa1.png | Bin 0 -> 32191 bytes assets/plsa2.png | Bin 0 -> 22340 bytes assets/qrsklatest.pdf | Bin 0 -> 436474 bytes assets/realdis.pdf | Bin 0 -> 698192 bytes assets/resources/1423graph.jpg | Bin 50282 -> 0 bytes assets/resources/dpmm.png | Bin 41295 -> 0 bytes assets/resources/fully-bayesian-mm.png | Bin 37129 -> 0 bytes assets/resources/hmm.png | Bin 57074 -> 0 bytes assets/resources/juggling.png | Bin 4802462 -> 0 bytes assets/resources/lda.png | Bin 43373 -> 0 bytes assets/resources/mixture-model.png | Bin 18380 -> 0 bytes assets/resources/plsa1.png | Bin 32191 -> 0 bytes assets/resources/plsa2.png | Bin 22340 -> 0 bytes assets/resources/qrsklatest.pdf | Bin 436474 -> 0 bytes assets/resources/realdis.pdf | Bin 698192 -> 0 bytes assets/resources/sl-vs-ml.png | Bin 150277 -> 0 bytes assets/resources/wangana.png | Bin 1447628 -> 0 bytes assets/resources/wangana_screenshot_index.png | Bin 90981 -> 0 bytes assets/sl-vs-ml.png | Bin 0 -> 150277 bytes assets/wangana.png | Bin 0 -> 1447628 bytes assets/wangana_screenshot_index.png | Bin 0 -> 90981 bytes .../2014-04-01-q-robinson-schensted-symmetry-paper.md | 2 +- ...2014-04-01-q-robinson-schensted-symmetry-paper.org | 2 +- posts/2015-04-02-juggling-skill-tree.md | 2 +- .../2016-10-13-q-robinson-schensted-knuth-polymer.md | 2 +- .../2016-10-13-q-robinson-schensted-knuth-polymer.org | 2 +- posts/2019-02-14-raise-your-elbo.md | 14 +++++++------- posts/2019-02-14-raise-your-elbo.org | 14 +++++++------- 35 files changed, 19 insertions(+), 19 deletions(-) create mode 100644 assets/1423graph.jpg create mode 100644 assets/dpmm.png create mode 100644 assets/fully-bayesian-mm.png create mode 100644 assets/hmm.png create mode 100644 assets/juggling.png create mode 100644 assets/lda.png create mode 100644 assets/mixture-model.png create mode 100644 assets/plsa1.png create mode 100644 assets/plsa2.png create mode 100644 assets/qrsklatest.pdf create mode 100644 assets/realdis.pdf delete mode 100644 assets/resources/1423graph.jpg delete mode 100644 assets/resources/dpmm.png delete mode 100644 assets/resources/fully-bayesian-mm.png delete mode 100644 assets/resources/hmm.png delete mode 100644 assets/resources/juggling.png delete mode 100644 assets/resources/lda.png delete mode 100644 assets/resources/mixture-model.png delete mode 100644 assets/resources/plsa1.png delete mode 100644 assets/resources/plsa2.png delete mode 100644 assets/resources/qrsklatest.pdf delete mode 100644 assets/resources/realdis.pdf delete mode 100644 assets/resources/sl-vs-ml.png delete mode 100644 assets/resources/wangana.png delete mode 100644 assets/resources/wangana_screenshot_index.png create mode 100644 assets/sl-vs-ml.png create mode 100644 assets/wangana.png create mode 100644 assets/wangana_screenshot_index.png diff --git a/assets/1423graph.jpg b/assets/1423graph.jpg new file mode 100644 index 0000000..1a6e739 Binary files /dev/null and b/assets/1423graph.jpg differ diff --git a/assets/dpmm.png b/assets/dpmm.png new file mode 100644 index 0000000..7d41b41 Binary files /dev/null and b/assets/dpmm.png differ diff --git a/assets/fully-bayesian-mm.png b/assets/fully-bayesian-mm.png new file mode 100644 index 0000000..5ce98c8 Binary files /dev/null and b/assets/fully-bayesian-mm.png differ diff --git a/assets/hmm.png b/assets/hmm.png new file mode 100644 index 0000000..da902cc Binary files /dev/null and b/assets/hmm.png differ diff --git a/assets/juggling.png b/assets/juggling.png new file mode 100644 index 0000000..18ec08b Binary files /dev/null and b/assets/juggling.png differ diff --git a/assets/lda.png b/assets/lda.png new file mode 100644 index 0000000..3eef014 Binary files /dev/null and b/assets/lda.png differ diff --git a/assets/mixture-model.png b/assets/mixture-model.png new file mode 100644 index 0000000..b0787a5 Binary files /dev/null and b/assets/mixture-model.png differ diff --git a/assets/plsa1.png b/assets/plsa1.png new file mode 100644 index 0000000..8fa81a6 Binary files /dev/null and b/assets/plsa1.png differ diff --git a/assets/plsa2.png b/assets/plsa2.png new file mode 100644 index 0000000..6b38829 Binary files /dev/null and b/assets/plsa2.png differ diff --git a/assets/qrsklatest.pdf b/assets/qrsklatest.pdf new file mode 100644 index 0000000..01a042b Binary files /dev/null and b/assets/qrsklatest.pdf differ diff --git a/assets/realdis.pdf b/assets/realdis.pdf new file mode 100644 index 0000000..92d9302 Binary files /dev/null and b/assets/realdis.pdf differ diff --git a/assets/resources/1423graph.jpg b/assets/resources/1423graph.jpg deleted file mode 100644 index 1a6e739..0000000 Binary files a/assets/resources/1423graph.jpg and /dev/null differ diff --git a/assets/resources/dpmm.png b/assets/resources/dpmm.png deleted file mode 100644 index 7d41b41..0000000 Binary files a/assets/resources/dpmm.png and /dev/null differ diff --git a/assets/resources/fully-bayesian-mm.png b/assets/resources/fully-bayesian-mm.png deleted file mode 100644 index 5ce98c8..0000000 Binary files a/assets/resources/fully-bayesian-mm.png and /dev/null differ diff --git a/assets/resources/hmm.png b/assets/resources/hmm.png deleted file mode 100644 index da902cc..0000000 Binary files a/assets/resources/hmm.png and /dev/null differ diff --git a/assets/resources/juggling.png b/assets/resources/juggling.png deleted file mode 100644 index 18ec08b..0000000 Binary files a/assets/resources/juggling.png and /dev/null differ diff --git a/assets/resources/lda.png b/assets/resources/lda.png deleted file mode 100644 index 3eef014..0000000 Binary files a/assets/resources/lda.png and /dev/null differ diff --git a/assets/resources/mixture-model.png b/assets/resources/mixture-model.png deleted file mode 100644 index b0787a5..0000000 Binary files a/assets/resources/mixture-model.png and /dev/null differ diff --git a/assets/resources/plsa1.png b/assets/resources/plsa1.png deleted file mode 100644 index 8fa81a6..0000000 Binary files a/assets/resources/plsa1.png and /dev/null differ diff --git a/assets/resources/plsa2.png b/assets/resources/plsa2.png deleted file mode 100644 index 6b38829..0000000 Binary files a/assets/resources/plsa2.png and /dev/null differ diff --git a/assets/resources/qrsklatest.pdf b/assets/resources/qrsklatest.pdf deleted file mode 100644 index 01a042b..0000000 Binary files a/assets/resources/qrsklatest.pdf and /dev/null differ diff --git a/assets/resources/realdis.pdf b/assets/resources/realdis.pdf deleted file mode 100644 index 92d9302..0000000 Binary files a/assets/resources/realdis.pdf and /dev/null differ diff --git a/assets/resources/sl-vs-ml.png b/assets/resources/sl-vs-ml.png deleted file mode 100644 index 11ad6a8..0000000 Binary files a/assets/resources/sl-vs-ml.png and /dev/null differ diff --git a/assets/resources/wangana.png b/assets/resources/wangana.png deleted file mode 100644 index b8457dc..0000000 Binary files a/assets/resources/wangana.png and /dev/null differ diff --git a/assets/resources/wangana_screenshot_index.png b/assets/resources/wangana_screenshot_index.png deleted file mode 100644 index ecaa8ac..0000000 Binary files a/assets/resources/wangana_screenshot_index.png and /dev/null differ diff --git a/assets/sl-vs-ml.png b/assets/sl-vs-ml.png new file mode 100644 index 0000000..11ad6a8 Binary files /dev/null and b/assets/sl-vs-ml.png differ diff --git a/assets/wangana.png b/assets/wangana.png new file mode 100644 index 0000000..b8457dc Binary files /dev/null and b/assets/wangana.png differ diff --git a/assets/wangana_screenshot_index.png b/assets/wangana_screenshot_index.png new file mode 100644 index 0000000..ecaa8ac Binary files /dev/null and b/assets/wangana_screenshot_index.png differ diff --git a/posts/2014-04-01-q-robinson-schensted-symmetry-paper.md b/posts/2014-04-01-q-robinson-schensted-symmetry-paper.md index 38874bb..a0432c8 100644 --- a/posts/2014-04-01-q-robinson-schensted-symmetry-paper.md +++ b/posts/2014-04-01-q-robinson-schensted-symmetry-paper.md @@ -13,7 +13,7 @@ generalisation of the growth diagram approach introduced by Fomin. This approach, which uses "growth graphs", can also be applied to a wider class of insertion algorithms which have a branching structure. -![Growth graph of q-RS for 1423](../assets/resources/1423graph.jpg) +![Growth graph of q-RS for 1423](../assets/1423graph.jpg) Above is the growth graph of the \\(q\\)-weighted Robinson-Schensted algorithm for the permutation \\({1 2 3 4\\choose1 4 2 3}\\). diff --git a/posts/2014-04-01-q-robinson-schensted-symmetry-paper.org b/posts/2014-04-01-q-robinson-schensted-symmetry-paper.org index b1c967d..7411a5d 100644 --- a/posts/2014-04-01-q-robinson-schensted-symmetry-paper.org +++ b/posts/2014-04-01-q-robinson-schensted-symmetry-paper.org @@ -10,7 +10,7 @@ approach, which uses "growth graphs", can also be applied to a wider class of insertion algorithms which have a branching structure. #+caption: Growth graph of q-RS for 1423 -[[../assets/resources/1423graph.jpg]] +[[../assets/1423graph.jpg]] Above is the growth graph of the \(q\)-weighted Robinson-Schensted algorithm for the permutation \({1 2 3 4\choose1 4 2 3}\). diff --git a/posts/2015-04-02-juggling-skill-tree.md b/posts/2015-04-02-juggling-skill-tree.md index 11eb377..7708988 100644 --- a/posts/2015-04-02-juggling-skill-tree.md +++ b/posts/2015-04-02-juggling-skill-tree.md @@ -30,4 +30,4 @@ have therefore written a script using [Python](http://python.org), difficulties, which is the leftmost column) from the Library of Juggling (click the image for the full size): -The juggling skill tree +The juggling skill tree diff --git a/posts/2016-10-13-q-robinson-schensted-knuth-polymer.md b/posts/2016-10-13-q-robinson-schensted-knuth-polymer.md index 4d31e37..4a5f793 100644 --- a/posts/2016-10-13-q-robinson-schensted-knuth-polymer.md +++ b/posts/2016-10-13-q-robinson-schensted-knuth-polymer.md @@ -12,7 +12,7 @@ This article is available at [arXiv](https://arxiv.org/abs/1610.03692). It seems to me that one difference between arXiv and Github is that on arXiv each preprint has a few versions only. In Github many projects have a "dev" branch hosting continuous updates, whereas the master branch is where the stable releases live. -[Here]({{ site.url }}/assets/resources/qrsklatest.pdf) is a "dev" version of the article, which I shall push to arXiv when it stablises. Below is the changelog. +[Here]({{ site.url }}/assets/qrsklatest.pdf) is a "dev" version of the article, which I shall push to arXiv when it stablises. Below is the changelog. * 2017-01-12: Typos and grammar, arXiv v2. * 2016-12-20: Added remarks on the geometric \\(q\\)-pushTASEP. Added remarks on the converse of the Burke property. Added natural language description of the \\(q\\)RSK. Fixed typos. diff --git a/posts/2016-10-13-q-robinson-schensted-knuth-polymer.org b/posts/2016-10-13-q-robinson-schensted-knuth-polymer.org index 93da639..206a4ab 100644 --- a/posts/2016-10-13-q-robinson-schensted-knuth-polymer.org +++ b/posts/2016-10-13-q-robinson-schensted-knuth-polymer.org @@ -28,7 +28,7 @@ few versions only. In Github many projects have a "dev" branch hosting continuous updates, whereas the master branch is where the stable releases live. -[[file:%7B%7B%20site.url%20%7D%7D/assets/resources/qrsklatest.pdf][Here]] +[[file:/assets/qrsklatest.pdf][Here]] is a "dev" version of the article, which I shall push to arXiv when it stablises. Below is the changelog. diff --git a/posts/2019-02-14-raise-your-elbo.md b/posts/2019-02-14-raise-your-elbo.md index 4080d0b..36bd364 100644 --- a/posts/2019-02-14-raise-your-elbo.md +++ b/posts/2019-02-14-raise-your-elbo.md @@ -138,7 +138,7 @@ $p(x_{1 : m}; \theta)$. Represented as a DAG (a.k.a the plate notations), the model looks like this: -![](/assets/resources/mixture-model.png){style="width:250px"} +![](/assets/mixture-model.png){style="width:250px"} where the boxes with $m$ mean repitition for $m$ times, since there $m$ indepdent pairs of $(x, z)$, and the same goes for $\eta$. @@ -298,7 +298,7 @@ $$p(d_i = u, x_i = w | z_i = k; \theta) = p(d_i ; \xi_k) p(x_i; \eta_k) = \xi_{k The model can be illustrated in the plate notations: -![](/assets/resources/plsa1.png){style="width:350px"} +![](/assets/plsa1.png){style="width:350px"} So the solution of the M-step is @@ -365,7 +365,7 @@ pLSA1, $(x | z = k) \sim \text{Cat}(\eta_{k, \cdot})$. Illustrated in the plate notations, pLSA2 is: -![](/assets/resources/plsa2.png){style="width:350px"} +![](/assets/plsa2.png){style="width:350px"} The computation is basically adding an index $\ell$ to the computation of SMM wherever applicable. @@ -411,7 +411,7 @@ $$p(z_{i1}) = \pi_{z_{i1}}$$ So the parameters are $\theta = (\pi, \xi, \eta)$. And HMM can be shown in plate notations as: -![](/assets/resources/hmm.png){style="width:350px"} +![](/assets/hmm.png){style="width:350px"} Now we apply EM to HMM, which is called the [Baum-Welch algorithm](https://en.wikipedia.org/wiki/Baum%E2%80%93Welch_algorithm). @@ -569,7 +569,7 @@ later in this section that the posterior $q(\eta_k)$ belongs to the same family as $p(\eta_k)$. Represented in a plate notations, a fully Bayesian mixture model looks like: -![](/assets/resources/fully-bayesian-mm.png){style="width:450px"} +![](/assets/fully-bayesian-mm.png){style="width:450px"} Given this structure we can write down the mean-field approximation: @@ -672,7 +672,7 @@ As the second example of fully Bayesian mixture models, Latent Dirichlet allocation (LDA) (Blei-Ng-Jordan 2003) is the fully Bayesian version of pLSA2, with the following plate notations: -![](/assets/resources/lda.png){style="width:450px"} +![](/assets/lda.png){style="width:450px"} It is the smoothed version in the paper. @@ -782,7 +782,7 @@ $$L(p, q) = \sum_{k = 1 : T} \mathbb E_{q(\theta_k)} \log {p(\theta_k) \over q(\ The plate notation of this model is: -![](/assets/resources/dpmm.png){style="width:450px"} +![](/assets/dpmm.png){style="width:450px"} As it turns out, the infinities can be tamed in this case. diff --git a/posts/2019-02-14-raise-your-elbo.org b/posts/2019-02-14-raise-your-elbo.org index f0de7d1..9cc79a7 100644 --- a/posts/2019-02-14-raise-your-elbo.org +++ b/posts/2019-02-14-raise-your-elbo.org @@ -137,7 +137,7 @@ $p(x_{1 : m}; \theta)$. Represented as a DAG (a.k.a the plate notations), the model looks like this: -[[/assets/resources/mixture-model.png]] +[[/assets/mixture-model.png]] where the boxes with $m$ mean repitition for $m$ times, since there $m$ indepdent pairs of $(x, z)$, and the same goes for $\eta$. @@ -309,7 +309,7 @@ $$p(d_i = u, x_i = w | z_i = k; \theta) = p(d_i ; \xi_k) p(x_i; \eta_k) = \xi_{k The model can be illustrated in the plate notations: -[[/assets/resources/plsa1.png]] +[[/assets/plsa1.png]] So the solution of the M-step is @@ -380,7 +380,7 @@ pLSA1, $(x | z = k) \sim \text{Cat}(\eta_{k, \cdot})$. Illustrated in the plate notations, pLSA2 is: -[[/assets/resources/plsa2.png]] +[[/assets/plsa2.png]] The computation is basically adding an index $\ell$ to the computation of SMM wherever applicable. @@ -429,7 +429,7 @@ $$p(z_{i1}) = \pi_{z_{i1}}$$ So the parameters are $\theta = (\pi, \xi, \eta)$. And HMM can be shown in plate notations as: -[[/assets/resources/hmm.png]] +[[/assets/hmm.png]] Now we apply EM to HMM, which is called the [[https://en.wikipedia.org/wiki/Baum%E2%80%93Welch_algorithm][Baum-Welch @@ -592,7 +592,7 @@ later in this section that the posterior $q(\eta_k)$ belongs to the same family as $p(\eta_k)$. Represented in a plate notations, a fully Bayesian mixture model looks like: -[[/assets/resources/fully-bayesian-mm.png]] +[[/assets/fully-bayesian-mm.png]] Given this structure we can write down the mean-field approximation: @@ -701,7 +701,7 @@ As the second example of fully Bayesian mixture models, Latent Dirichlet allocation (LDA) (Blei-Ng-Jordan 2003) is the fully Bayesian version of pLSA2, with the following plate notations: -[[/assets/resources/lda.png]] +[[/assets/lda.png]] It is the smoothed version in the paper. @@ -813,7 +813,7 @@ $$L(p, q) = \sum_{k = 1 : T} \mathbb E_{q(\theta_k)} \log {p(\theta_k) \over q(\ The plate notation of this model is: -[[/assets/resources/dpmm.png]] +[[/assets/dpmm.png]] As it turns out, the infinities can be tamed in this case. -- cgit v1.2.3