Skip to content

Commit 7c19b10

Browse files
xames3xames3
committed
docs: upstream content about machine learning
Signed-off-by: Akshay Mestry <[email protected]>
1 parent 0bf7818 commit 7c19b10

File tree

5 files changed

+203
-2
lines changed

5 files changed

+203
-2
lines changed

‎docs/source/assets/it-works-why.jpg

40.8 KB
Loading

‎docs/source/index.rst

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
.. Author: Akshay Mestry <[email protected]>
22
.. Created on: Saturday, February 22, 2025
3-
.. Last updated on: Sunday, April 20 2025
3+
.. Last updated on: Saturday, May 03 2025
44
55
:orphan:
66
:og:title: Studying, Mentorship, And Resourceful Teaching
@@ -21,7 +21,7 @@ My Studying, Mentorship, And Resourceful Teaching Journey
2121
:avatar: https://avatars.githubusercontent.com/u/90549089?v=4
2222
:github: https://github.com/xames3
2323
:linkedin: https://linkedin.com/in/xames3
24-
:timestamp: Feb 23, 2025
24+
:timestamp: May 03, 2025
2525

2626
.. rst-class:: lead
2727

@@ -153,6 +153,7 @@ sharing ideas, and helping others grow alongside me.
153153
:caption: Teaching
154154
:hidden:
155155

156+
learning-out-loud/index
156157
course-codex/index
157158

158159
.. toctree::

‎docs/source/learning-out-loud/index.rst

Lines changed: 60 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,60 @@
1+
.. Author: Akshay Mestry <[email protected]>
2+
.. Created on: Friday, April 25 2025
3+
.. Last updated on: Saturday, May 03 2025
4+
5+
:og:title: Learning Out Loud
6+
:og:description: When one teaches, two learn...
7+
:og:type: article
8+
9+
.. _learning-out-loud-index:
10+
11+
===============================================================================
12+
Learning Out Loud
13+
===============================================================================
14+
15+
.. author::
16+
:name: Akshay Mestry
17+
:email: [email protected]
18+
:about: National Louis University
19+
:avatar: https://avatars.githubusercontent.com/u/90549089?v=4
20+
:github: https://github.com/xames3
21+
:linkedin: https://linkedin.com/in/xames3
22+
:timestamp: Apr 25, 2025
23+
24+
This article isn't just a space to stash my AI and machine learning notes,
25+
it's a living reflection of how I've come to understand these ideas, and how I
26+
continue to wrestle with them in public. Ever since I first started getting
27+
into machine learning in 2018, I've been fascinated not just by what these
28+
"models" can do, but by how we come to understand them. I've made my fair
29+
share of mistakes, read the same paper five times just to make sense of one
30+
paragraph, and written buggy models that taught me more than any lecture
31+
could. But somewhere in all that trial and error, I also found clarity |dash|
32+
the kind that sticks when you try to teach something to someone else.
33+
34+
This section, Learning Out Loud (LOL), is my humble attempt to do just that.
35+
It's not a course. It's not a tutorial series in the traditional sense. It's
36+
a place where I unpack, explain, and reflect on concepts in machine learning
37+
and AI. If :doc:`../course-codex/index` is my attempt to design the courses I
38+
wish I'd been taught, then this is where I actually teach them. Out loud. As I
39+
think. As I learn. Slowly, thoughtfully, and with the hope that it helps
40+
someone else connect the dots. If you're reading this (thank you by the way),
41+
I'm not here to dazzle you with complexity or gatekeep behind jargon. My aim
42+
is to build understanding from the ground up, to offer practical explanations
43+
grounded in real engineering work, and to make space for thinking out loud,
44+
especially when things are murky.
45+
46+
Some articles will feel like quiet lectures. Others will read more like lab
47+
notes or rants from a whiteboard session. All of them are teaching moments for
48+
you, and for me. These are not polished lectures or formal research. They're
49+
more like annotated ideas, teachable moments, and deep dives written for
50+
students, engineers, and the endlessly curious.
51+
52+
So whether you're just starting out, teaching others, or knee-deep in your own
53+
experiments, I hope you find something here that makes you pause and say, "Ah,
54+
now that makes sense." And if not... that's alright too. Learning out loud
55+
means we get to revisit things as many times as we need.
56+
57+
.. toctree::
58+
:hidden:
59+
60+
ml-explained/index

‎docs/source/learning-out-loud/ml-explained/index.rst

Lines changed: 46 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,46 @@
1+
.. Author: Akshay Mestry <[email protected]>
2+
.. Created on: Friday, April 25 2025
3+
.. Last updated on: Saturday, May 03 2025
4+
5+
:og:title: ML Explained
6+
:og:description: A narrative series that walks through the foundations of
7+
Machine Learning from first principles.
8+
:og:type: article
9+
10+
.. _ml-explained-index:
11+
12+
===============================================================================
13+
ML Explained
14+
===============================================================================
15+
16+
.. author::
17+
:name: Akshay Mestry
18+
:email: [email protected]
19+
:about: National Louis University
20+
:avatar: https://avatars.githubusercontent.com/u/90549089?v=4
21+
:github: https://github.com/xames3
22+
:linkedin: https://linkedin.com/in/xames3
23+
:timestamp: May 03, 2025
24+
25+
This corner of the internet is the place where I attempt to teach Machine
26+
Learning the way I wish I'd first encountered it... slowly, clearly, and with
27+
context that sticks. If you've ever googled, "machine learning" and landed on a
28+
sea of buzzwords, equations, or flowcharts that left you more confused than
29+
enlightened; welcome! You're not alone. I've been there too, and it's part of
30+
why I'm writing this series.
31+
32+
The people that know me well, know that I do not have a good attention span. So
33+
I'll keep things short. This explainer series is not going to be a tutorial or
34+
a crash course. It's more like a conversation. A walk-through of the key
35+
ideas, the questions worth asking, and the principles you actually need to
36+
understand before you care about the accuracy scores or the latest Transformer
37+
papers out there. Like I said earlier, I'd like to keep things short but, not
38+
incomplete. Each article will be hopefully short, focused, and designed to
39+
build up your intuition over time.
40+
41+
This is not about "Mastering ML in X days". It's about learning out loud and
42+
learning well.
43+
44+
.. toctree::
45+
46+
ml101

‎docs/source/learning-out-loud/ml-explained/ml101.rst

Lines changed: 94 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,94 @@
1+
.. Author: Akshay Mestry <[email protected]>
2+
.. Created on: Friday, April 25 2025
3+
.. Last updated on: Saturday, May 03 2025
4+
5+
:og:title: ML101
6+
:og:description: Understanding learning as function approximation, not magic.
7+
:og:type: article
8+
9+
.. _ml101:
10+
11+
===============================================================================
12+
ML101
13+
===============================================================================
14+
15+
.. author::
16+
:name: Akshay Mestry
17+
:email: [email protected]
18+
:about: National Louis University
19+
:avatar: https://avatars.githubusercontent.com/u/90549089?v=4
20+
:github: https://github.com/xames3
21+
:linkedin: https://linkedin.com/in/xames3
22+
:timestamp: May 03, 2025
23+
24+
.. rst-class:: lead
25+
26+
This isn't a crash course. There's no "ultimate guide" here, no promise to
27+
make you an expert over a weekend
28+
29+
To be fair, this doesn't really need explaining. If you're here, chances are
30+
you already have some sense of what Machine Learning is, or at least you feel
31+
you ought to. I'm not about to hand you some ground-breaking new definition.
32+
What I do hope to offer is clarity, to dispel some of the haze, the
33+
half-truths, and the misguided metaphors that surround Machine Learning. Not
34+
just in the press or on LinkedIn slides, but also in the way it's taught,
35+
explained, and even practised by fellow professionals. You see, Machine
36+
Learning is often treated like a black box that "just works", a clever
37+
contraption that learns because it's somehow intelligent. You collect some
38+
data, feed it into a model, twiddle a few parameters, and voilà, it learns.
39+
Except it doesn't, not in the way you might think. The word "learning" here is
40+
doing rather a lot of heavy lifting.
41+
42+
.. image:: ../../assets/it-works-why.jpg
43+
:alt: It works... why meme
44+
45+
The issue isn't that people are unaware of Machine Learning, it's that we
46+
seldom pause to understand it deeply. The explanations on offer, blog posts,
47+
conference talks, YouTube tutorials, often skip the foundations. We learn
48+
**how** long before we ever touch **why**. And without the **why**, Machine
49+
Learning becomes little more than glorified pattern recognition wrapped in
50+
prestige.
51+
52+
.. _a-new-kind-of-programming:
53+
54+
-------------------------------------------------------------------------------
55+
A New Kind of Programming?
56+
-------------------------------------------------------------------------------
57+
58+
Machine Learning, at its essence, is a different philosophy of programming.
59+
Before I get ahead of myself, let's try to understand what programming is. To
60+
keep things simple, programming is an "art" of writing programs... duh! I mean
61+
there's more to that like development, testing, integration, etc. But what it
62+
boils down to is writing a code or program which is nothing but a fancy way of
63+
writing instructions which the computer (machine) will follow. So, basically
64+
a program is a set of instructions or when you say I'm a programmer, I write
65+
instructions for the computer (machine) to follow along. If you've taken any
66+
Machine Learning class before, you might've come across this definition...
67+
68+
.. epigraph::
69+
70+
Machine Learning is the field of study that gives computers the ability to
71+
learn without being explicitly programmed
72+
73+
-- Arthur Samuel, 1959
74+
75+
In the classical approach, we often write explicit instructions, handcrafted
76+
rules, conditional logic, or, to keep it precise, programs. The machine doesn't
77+
think. It simply obeys or follows those rules or instructions.
78+
79+
Machine Learning flips this paradigm.
80+
81+
Instead of coding or programming the logic ourselves, we supply the machine
82+
(computer) with examples. And I mean a lot of them. By the way, these examples
83+
are inputs and the desired outputs. We then ask the algorithm (another
84+
program) to infer the rules or instructions that connect the two. That's the
85+
entire conceit. We don't program the rules. We let the machine learn them. You
86+
don't tell the system what to do; you show it what has been done and let it
87+
infer the rest.
88+
89+
And just because you let the "system" or "machine" learn, it
90+
doesn't mean you don't do anything. Learning is a process. It's not unlike
91+
teaching a child to ride a bicycle. You don't explain Newtonian mechanics or
92+
angular momentum. You run alongside them, steady the seat, and let them wobble.
93+
The learning comes along through doing. Like I said, its a process. The rules
94+
emerge from experience.

0 commit comments

Comments
 (0)