Update deep-learning.html
Browse files- deep-learning.html +24 -1
deep-learning.html
CHANGED
|
@@ -73,7 +73,30 @@
|
|
| 73 |
</div>
|
| 74 |
</div>
|
| 75 |
</section>
|
| 76 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
<section style="margin-top: 40px;">
|
| 78 |
<h2>Methodologies & Optimization Techniques</h2>
|
| 79 |
<div class="methodology-box">
|
|
|
|
| 73 |
</div>
|
| 74 |
</div>
|
| 75 |
</section>
|
| 76 |
+
<section style="margin-top: 40px; border: 1px solid #334155; padding: 25px; border-radius: 12px; background: linear-gradient(180deg, #161e2e, #0f172a);">
|
| 77 |
+
<h2 style="color: #ef4444;">The Hardware Catalyst: From Silicon to Neural Engines</h2>
|
| 78 |
+
<p>AI reached a "level up" not just because of better code, but because we changed the physical way computers think.</p>
|
| 79 |
+
|
| 80 |
+
<div class="tech-grid">
|
| 81 |
+
<div class="tech-item">
|
| 82 |
+
<h4>1. The GPU Revolution (Nvidia Shift)</h4>
|
| 83 |
+
<p><strong>The Move:</strong> Moving from CPU to GPU. While a CPU handles a few complex tasks in a row (Serial), a GPU handles thousands of simple math tasks at once (Parallel).</p>
|
| 84 |
+
<p><em>Why it matters:</em> Neural networks are just massive matrices of multiplication. GPUs can do billions of these per second.</p>
|
| 85 |
+
</div>
|
| 86 |
+
<div class="tech-item">
|
| 87 |
+
<h4>2. CUDA & Software Abstraction</h4>
|
| 88 |
+
<p><strong>The Move:</strong> Nvidia’s CUDA allowed researchers to write C++ code directly for the GPU. This turned a "video card" into a general-purpose AI brain.</p>
|
| 89 |
+
</div>
|
| 90 |
+
<div class="tech-item">
|
| 91 |
+
<h4>3. TPU (Tensor Processing Units)</h4>
|
| 92 |
+
<p><strong>The Move:</strong> Google developed ASICs (Application-Specific Integrated Circuits) designed <em>specifically</em> for the matrix math used in AI, stripping away everything a computer doesn't need for neural nets.</p>
|
| 93 |
+
</div>
|
| 94 |
+
<div class="tech-item">
|
| 95 |
+
<h4>4. HBM (High Bandwidth Memory)</h4>
|
| 96 |
+
<p><strong>The Move:</strong> The bottleneck wasn't just calculation speed; it was moving data to the chip. HBM allowed "stacks" of memory to sit right next to the processor, providing the speed needed for LLMs.</p>
|
| 97 |
+
</div>
|
| 98 |
+
</div>
|
| 99 |
+
</section>
|
| 100 |
<section style="margin-top: 40px;">
|
| 101 |
<h2>Methodologies & Optimization Techniques</h2>
|
| 102 |
<div class="methodology-box">
|