Skip to content

Commit e013f7f

Browse files
mhjensenmhjensen
committed
added videos
1 parent de454de commit e013f7f

File tree

8 files changed

+97
-75
lines changed

8 files changed

+97
-75
lines changed

‎doc/pub/week7/html/week7-bs.html‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -190,9 +190,9 @@ <h2 id="plans-for-the-week-of-march-3-7" class="anchor">Plans for the week of Ma
190190
<ol>
191191
<li> Reminder on basics of recurrent neural networks (RNNs)</li>
192192
<li> Mathematics of RNNs</li>
193-
<li> Writing our own codes for RNNs
194-
<!-- o <a href="https://youtu.be/VkQGq84ml_0" target="_self">Video of lecture</a> -->
195-
<!-- o <a href="https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesFebruary27.pdf" target="_self">Whiteboard notes</a> --></li>
193+
<li> Writing our own codes for RNNs</li>
194+
<li> <a href="https://youtu.be/MeYh5rGIRBM" target="_self">Video of lecture</a></li>
195+
<li> "Whiteboard notes":"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2025/NotesMarch6.pdf</li>
196196
<li> Reading recommendations:
197197
<ol type="a"></li>
198198
<li> Goodfellow, Bengio and Courville's chapter 10 from <a href="https://www.deeplearningbook.org/" target="_self">Deep Learning</a></li>

‎doc/pub/week7/html/week7-reveal.html‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -200,9 +200,9 @@ <h2 id="plans-for-the-week-of-march-3-7">Plans for the week of March 3-7 </h2>
200200
<ol>
201201
<p><li> Reminder on basics of recurrent neural networks (RNNs)</li>
202202
<p><li> Mathematics of RNNs</li>
203-
<p><li> Writing our own codes for RNNs
204-
<!-- o <a href="https://youtu.be/VkQGq84ml_0" target="_blank">Video of lecture</a> -->
205-
<!-- o <a href="https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesFebruary27.pdf" target="_blank">Whiteboard notes</a> --></li>
203+
<p><li> Writing our own codes for RNNs</li>
204+
<p><li> <a href="https://youtu.be/MeYh5rGIRBM" target="_blank">Video of lecture</a></li>
205+
<p><li> "Whiteboard notes":"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2025/NotesMarch6.pdf</li>
206206
<p><li> Reading recommendations:
207207
<ol type="a"></li>
208208
<p><li> Goodfellow, Bengio and Courville's chapter 10 from <a href="https://www.deeplearningbook.org/" target="_blank">Deep Learning</a></li>

‎doc/pub/week7/html/week7-solarized.html‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -172,9 +172,9 @@ <h2 id="plans-for-the-week-of-march-3-7">Plans for the week of March 3-7 </h2>
172172
<ol>
173173
<li> Reminder on basics of recurrent neural networks (RNNs)</li>
174174
<li> Mathematics of RNNs</li>
175-
<li> Writing our own codes for RNNs
176-
<!-- o <a href="https://youtu.be/VkQGq84ml_0" target="_blank">Video of lecture</a> -->
177-
<!-- o <a href="https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesFebruary27.pdf" target="_blank">Whiteboard notes</a> --></li>
175+
<li> Writing our own codes for RNNs</li>
176+
<li> <a href="https://youtu.be/MeYh5rGIRBM" target="_blank">Video of lecture</a></li>
177+
<li> "Whiteboard notes":"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2025/NotesMarch6.pdf</li>
178178
<li> Reading recommendations:
179179
<ol type="a"></li>
180180
<li> Goodfellow, Bengio and Courville's chapter 10 from <a href="https://www.deeplearningbook.org/" target="_blank">Deep Learning</a></li>

‎doc/pub/week7/html/week7.html‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -249,9 +249,9 @@ <h2 id="plans-for-the-week-of-march-3-7">Plans for the week of March 3-7 </h2>
249249
<ol>
250250
<li> Reminder on basics of recurrent neural networks (RNNs)</li>
251251
<li> Mathematics of RNNs</li>
252-
<li> Writing our own codes for RNNs
253-
<!-- o <a href="https://youtu.be/VkQGq84ml_0" target="_blank">Video of lecture</a> -->
254-
<!-- o <a href="https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesFebruary27.pdf" target="_blank">Whiteboard notes</a> --></li>
252+
<li> Writing our own codes for RNNs</li>
253+
<li> <a href="https://youtu.be/MeYh5rGIRBM" target="_blank">Video of lecture</a></li>
254+
<li> "Whiteboard notes":"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2025/NotesMarch6.pdf</li>
255255
<li> Reading recommendations:
256256
<ol type="a"></li>
257257
<li> Goodfellow, Bengio and Courville's chapter 10 from <a href="https://www.deeplearningbook.org/" target="_blank">Deep Learning</a></li>

‎doc/pub/week7/ipynb/ipynb-week7-src.tar.gz‎

0 Bytes
Binary file not shown.

‎doc/pub/week7/ipynb/week7.ipynb‎

Lines changed: 83 additions & 61 deletions
Original file line numberDiff line numberDiff line change
@@ -2,8 +2,10 @@
22
"cells": [
33
{
44
"cell_type": "markdown",
5-
"id": "e00f1a44",
6-
"metadata": {},
5+
"id": "2e9df165",
6+
"metadata": {
7+
"editable": true
8+
},
79
"source": [
810
"<!-- HTML file automatically generated from DocOnce source (https://github.com/doconce/doconce/)\n",
911
"doconce format html week7.do.txt --no_mako -->\n",
@@ -12,8 +14,10 @@
1214
},
1315
{
1416
"cell_type": "markdown",
15-
"id": "0521f5b9",
16-
"metadata": {},
17+
"id": "8d2e3ba3",
18+
"metadata": {
19+
"editable": true
20+
},
1721
"source": [
1822
"# Advanced machine learning and data analysis for the physical sciences\n",
1923
"**Morten Hjorth-Jensen**, Department of Physics and Center for Computing in Science Education, University of Oslo, Norway\n",
@@ -23,8 +27,10 @@
2327
},
2428
{
2529
"cell_type": "markdown",
26-
"id": "794145b7",
27-
"metadata": {},
30+
"id": "60bcfd05",
31+
"metadata": {
32+
"editable": true
33+
},
2834
"source": [
2935
"## Plans for the week of March 3-7\n",
3036
"\n",
@@ -33,10 +39,12 @@
3339
"2. Mathematics of RNNs\n",
3440
"\n",
3541
"3. Writing our own codes for RNNs\n",
36-
"<!-- o [Video of lecture](https://youtu.be/VkQGq84ml_0) -->\n",
37-
"<!-- o [Whiteboard notes](https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesFebruary27.pdf) -->\n",
3842
"\n",
39-
"4. Reading recommendations:\n",
43+
"4. [Video of lecture](https://youtu.be/MeYh5rGIRBM)\n",
44+
"\n",
45+
"5. \"Whiteboard notes\":\"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2025/NotesMarch6.pdf\n",
46+
"\n",
47+
"6. Reading recommendations:\n",
4048
"\n",
4149
"a. Goodfellow, Bengio and Courville's chapter 10 from [Deep Learning](https://www.deeplearningbook.org/)\n",
4250
"\n",
@@ -49,8 +57,10 @@
4957
},
5058
{
5159
"cell_type": "markdown",
52-
"id": "e5e42fd9",
53-
"metadata": {},
60+
"id": "0875ad23",
61+
"metadata": {
62+
"editable": true
63+
},
5464
"source": [
5565
"## What is a recurrent NN?\n",
5666
"\n",
@@ -68,8 +78,10 @@
6878
},
6979
{
7080
"cell_type": "markdown",
71-
"id": "8ea5b91a",
72-
"metadata": {},
81+
"id": "4b3bc6ff",
82+
"metadata": {
83+
"editable": true
84+
},
7385
"source": [
7486
"## Why RNNs?\n",
7587
"\n",
@@ -88,8 +100,10 @@
88100
},
89101
{
90102
"cell_type": "markdown",
91-
"id": "6dadec9c",
92-
"metadata": {},
103+
"id": "1a5d8045",
104+
"metadata": {
105+
"editable": true
106+
},
93107
"source": [
94108
"## Basic layout, [Figures from Sebastian Rashcka et al, Machine learning with Sickit-Learn and PyTorch](https://sebastianraschka.com/blog/2022/ml-pytorch-book.html)\n",
95109
"\n",
@@ -102,8 +116,10 @@
102116
},
103117
{
104118
"cell_type": "markdown",
105-
"id": "f69c94e7",
106-
"metadata": {},
119+
"id": "b8620ba2",
120+
"metadata": {
121+
"editable": true
122+
},
107123
"source": [
108124
"## RNNs in more detail\n",
109125
"\n",
@@ -116,8 +132,10 @@
116132
},
117133
{
118134
"cell_type": "markdown",
119-
"id": "3e66ff38",
120-
"metadata": {},
135+
"id": "a4204887",
136+
"metadata": {
137+
"editable": true
138+
},
121139
"source": [
122140
"## RNNs in more detail, part 2\n",
123141
"\n",
@@ -130,8 +148,10 @@
130148
},
131149
{
132150
"cell_type": "markdown",
133-
"id": "a485dcbf",
134-
"metadata": {},
151+
"id": "6303caf2",
152+
"metadata": {
153+
"editable": true
154+
},
135155
"source": [
136156
"## RNNs in more detail, part 3\n",
137157
"\n",
@@ -144,8 +164,10 @@
144164
},
145165
{
146166
"cell_type": "markdown",
147-
"id": "ea3e2dd7",
148-
"metadata": {},
167+
"id": "579a3e09",
168+
"metadata": {
169+
"editable": true
170+
},
149171
"source": [
150172
"## RNNs in more detail, part 4\n",
151173
"\n",
@@ -158,8 +180,10 @@
158180
},
159181
{
160182
"cell_type": "markdown",
161-
"id": "5eed88f5",
162-
"metadata": {},
183+
"id": "8a879d38",
184+
"metadata": {
185+
"editable": true
186+
},
163187
"source": [
164188
"## RNNs in more detail, part 5\n",
165189
"\n",
@@ -172,8 +196,10 @@
172196
},
173197
{
174198
"cell_type": "markdown",
175-
"id": "f4dd3d82",
176-
"metadata": {},
199+
"id": "7930ce08",
200+
"metadata": {
201+
"editable": true
202+
},
177203
"source": [
178204
"## RNNs in more detail, part 6\n",
179205
"\n",
@@ -186,8 +212,10 @@
186212
},
187213
{
188214
"cell_type": "markdown",
189-
"id": "f2f57990",
190-
"metadata": {},
215+
"id": "850d2d15",
216+
"metadata": {
217+
"editable": true
218+
},
191219
"source": [
192220
"## RNNs in more detail, part 7\n",
193221
"\n",
@@ -200,8 +228,10 @@
200228
},
201229
{
202230
"cell_type": "markdown",
203-
"id": "f86b6161",
204-
"metadata": {},
231+
"id": "d5fce222",
232+
"metadata": {
233+
"editable": true
234+
},
205235
"source": [
206236
"## Backpropagation through time\n",
207237
"\n",
@@ -219,8 +249,10 @@
219249
},
220250
{
221251
"cell_type": "markdown",
222-
"id": "40adbc7d",
223-
"metadata": {},
252+
"id": "bdf76896",
253+
"metadata": {
254+
"editable": true
255+
},
224256
"source": [
225257
"## The backward pass is linear\n",
226258
"\n",
@@ -236,8 +268,10 @@
236268
},
237269
{
238270
"cell_type": "markdown",
239-
"id": "bc9082ba",
240-
"metadata": {},
271+
"id": "31005fd6",
272+
"metadata": {
273+
"editable": true
274+
},
241275
"source": [
242276
"## The problem of exploding or vanishing gradients\n",
243277
"* What happens to the magnitude of the gradients as we backpropagate through many layers?\n",
@@ -259,8 +293,10 @@
259293
},
260294
{
261295
"cell_type": "markdown",
262-
"id": "9f380618",
263-
"metadata": {},
296+
"id": "6ee2e70b",
297+
"metadata": {
298+
"editable": true
299+
},
264300
"source": [
265301
"## The mathematics of RNNs, the basic architecture\n",
266302
"\n",
@@ -269,11 +305,13 @@
269305
},
270306
{
271307
"cell_type": "markdown",
272-
"id": "2b6f4cbd",
273-
"metadata": {},
308+
"id": "bd5842f9",
309+
"metadata": {
310+
"editable": true
311+
},
274312
"source": [
275313
"## Four effective ways to learn an RNN and preparing for next week\n",
276-
"1. Long Short Term Memory Make the RNN out of little modules that are designed to remember values for a long time. We discuss this next week.\n",
314+
"1. Long Short Term Memory Make the RNN out of little modules that are designed to remember values for a long time.\n",
277315
"\n",
278316
"2. Hessian Free Optimization: Deal with the vanishing gradients problem by using a fancy optimizer that can detect directions with a tiny gradient but even smaller curvature.\n",
279317
"\n",
@@ -284,8 +322,10 @@
284322
},
285323
{
286324
"cell_type": "markdown",
287-
"id": "ba57c26d",
288-
"metadata": {},
325+
"id": "3406a1ad",
326+
"metadata": {
327+
"editable": true
328+
},
289329
"source": [
290330
"## Long Short Term Memory (LSTM)\n",
291331
"\n",
@@ -305,25 +345,7 @@
305345
]
306346
}
307347
],
308-
"metadata": {
309-
"kernelspec": {
310-
"display_name": "Python 3 (ipykernel)",
311-
"language": "python",
312-
"name": "python3"
313-
},
314-
"language_info": {
315-
"codemirror_mode": {
316-
"name": "ipython",
317-
"version": 3
318-
},
319-
"file_extension": ".py",
320-
"mimetype": "text/x-python",
321-
"name": "python",
322-
"nbconvert_exporter": "python",
323-
"pygments_lexer": "ipython3",
324-
"version": "3.9.15"
325-
}
326-
},
348+
"metadata": {},
327349
"nbformat": 4,
328350
"nbformat_minor": 5
329351
}

‎doc/pub/week7/pdf/week7.pdf‎

535 Bytes
Binary file not shown.

‎doc/src/week7/week7.do.txt‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -9,8 +9,8 @@ DATE: March 6
99
o Reminder on basics of recurrent neural networks (RNNs)
1010
o Mathematics of RNNs
1111
o Writing our own codes for RNNs
12-
#o "Video of lecture":"https://youtu.be/VkQGq84ml_0"
13-
#o "Whiteboard notes":"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesFebruary27.pdf"
12+
o "Video of lecture":"https://youtu.be/MeYh5rGIRBM"
13+
o "Whiteboard notes":"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2025/NotesMarch6.pdf
1414
o Reading recommendations:
1515
o Goodfellow, Bengio and Courville's chapter 10 from "Deep Learning":"https://www.deeplearningbook.org/"
1616
o "Sebastian Rashcka et al, chapter 15, Machine learning with Sickit-Learn and PyTorch":"https://sebastianraschka.com/blog/2022/ml-pytorch-book.html"

0 commit comments

Comments
 (0)