DEV Community

Machine Learning

A branch of artificial intelligence (AI) and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
From Attention to 3-Bit Caches: How LLMs Learned to Remember Less (MHA, GQA, MLA, MoE, Quantisation Explained)

From Attention to 3-Bit Caches: How LLMs Learned to Remember Less (MHA, GQA, MLA, MoE, Quantisation Explained)

7
Comments
22 min read
LLMs Aren’t the Only Game in Town: A Beginner’s Guide to JEPA, Concept Models and Decision Models

LLMs Aren’t the Only Game in Town: A Beginner’s Guide to JEPA, Concept Models and Decision Models

5
Comments
11 min read
OpenAI Failed at the Checkout. Now It Wants the Conversation Before It

OpenAI Failed at the Checkout. Now It Wants the Conversation Before It

Comments 6
2 min read
Passive Rangefinding Without LiDAR: Depth-From-Defocus and Optical Physics in Pure NumPy

Passive Rangefinding Without LiDAR: Depth-From-Defocus and Optical Physics in Pure NumPy

2
Comments
14 min read
Run Gemma 4 31B-it Uncensored on a 12GB GPU: Complete Local Vision & 256K Context Guide with llama.cpp, DeepSeek Harness & Hermes Agent

Run Gemma 4 31B-it Uncensored on a 12GB GPU: Complete Local Vision & 256K Context Guide with llama.cpp, DeepSeek Harness & Hermes Agent

Comments
14 min read
Parsing a whole PDF in one request with Unlimited OCR 2: the crop_mode setting and a silent cut

Parsing a whole PDF in one request with Unlimited OCR 2: the crop_mode setting and a silent cut

Comments
4 min read
Five People Just Won $500 Each for Testing AI Properly. I Read All Five Benchmarks.

Five People Just Won $500 Each for Testing AI Properly. I Read All Five Benchmarks.

1
Comments
4 min read
I built a free Windows (UEFI) & Linux GUI to unlock RTX 40/30/20 Tensor Cores (2.01x FP16/BF16/FP8 on my RTX 4070 Ti)

I built a free Windows (UEFI) & Linux GUI to unlock RTX 40/30/20 Tensor Cores (2.01x FP16/BF16/FP8 on my RTX 4070 Ti)

1
Comments
3 min read
MuJoCo for Developers: A Practical Guide to Google DeepMind's Physics Engine

MuJoCo for Developers: A Practical Guide to Google DeepMind's Physics Engine

5
Comments
8 min read
Building a Model Monitoring System for ML Drift Detection

Building a Model Monitoring System for ML Drift Detection

Comments
5 min read
Seven Neurons and a Trigger String: What Microsoft's Lazarus Backdoor Actually Proves

Seven Neurons and a Trigger String: What Microsoft's Lazarus Backdoor Actually Proves

Comments
5 min read
Certified kernels instead of autotuning on your machine

Certified kernels instead of autotuning on your machine

Comments
4 min read
A useful vocal isolate is not a karaoke track

A useful vocal isolate is not a karaoke track

Comments
1 min read
In-browser Basic Pitch wants a single line, not the full mix

In-browser Basic Pitch wants a single line, not the full mix

Comments
2 min read
Diluting Poisoned Data Made the Backdoor Stronger: Anthropic's 'Homeopathic Learning' Explained

Diluting Poisoned Data Made the Backdoor Stronger: Anthropic's 'Homeopathic Learning' Explained

Comments
4 min read
👋 Sign in for the ability to sort posts by relevant, latest, or top.