Second Challenge

Now, let's build your very own AI.

AI is probability. So in the end, building it means making a program that computes probabilities. You don't need to know a programming language. Here you firmly grasp the concept, then just tell Laria how you want to build it.

Code it yourself in the AI Lab

Key Concept

Based on the vast amount of information on the internet, it collects and analyzes words one by one, and precomputes the probability of which word comes after a given word.

When you ask a question, it assembles words the way it stored them, calculates which word should come next, and strings together the highest-scoring candidates to form a sentence.

A Plan That Leads to Real Development

Start small like a toy, understand the principles, then grow it bigger over time.

Step 1

Gather words

Collect sentences about the topic you want to learn. Just 10 sentences is enough to start.

Step 2

Count next words

Make a table of how many times each word follows another word.

Step 3

Compute probability scores

Divide the count by the total number of candidates to get the probability score for the next word.

Step 4

Link the top candidate

String together the words with the highest probability, and you get a tiny sentence generator.

Step 5

Improve it your way

Add better sentences, more data, and different selection rules to grow it into your own AI.

A Tiny Probability Calculator

Change the corpus below and the next-word candidates and their probabilities change instantly. This is the smallest form of "your own AI."

Training sentences Starting sentence

Next-word candidates 21 words learned

다음 50%

appears 1×

확률입니다 50%

appears 1×

Generated sentence 인공지능은 다음 단어를 예측합니다 하늘은 맑습니다 학생은 질문합니다 라리아는 쉽게 설명합니다 라리아는 쉽게 설명합니다 라리아는

A Development Request to Hand to Laria

Send this request to the Laria chatbot on the right, and you can design together the steps to expand the little probability calculator you just made into real code.

라리아, 내가 직접 작은 인공지능을 만들고 싶어.

목표: 다음 단어를 확률로 예측하는 계산기
훈련 문장: 인공지능은 확률입니다. 인공지능은 다음 단어를 예측합니다. 하늘은 파랗습니다. 하늘은 맑습니다. 학생은 질문합니다. 라리아는 쉽게 설명합니다. 라리아는 재미있게 설명합니다. 확률은 후보를 비교합니다.
시작 문장: 인공지능은

1단계부터 아주 쉽게 설명하고, 내가 어떤 데이터를 더 넣으면 좋은지 질문해 줘.