nowfound

Life & fun · March 7, 2018

SV

Solving visual analogy puzzles with Deep Learning

Bongard problmes are named after their inventor, Soviet computer scientist Mikhail Bongard, who was working on pattern recognition in the 1960s. He designed 100 of this puzzles, to be a good benchmark for pattern recognition abilities, and they seem to be challenging for both people and algorithms.

Solving visual analogy puzzles with Deep Learning launched on March 7, 2018 with 60 votes, #722 of 2,954 launches that month and more than 90% of that year's launches. The site was still online in September 2026, 8 years on.

Visit k10v.github.io ↗Alternativestop 24% of March 2018

In plain words

This project applies deep learning to solve Bongard problems, visual puzzles created by Soviet computer scientist Mikhail Bongard in the 1960s. Users are presented with six images following one pattern and six following a different pattern, then must identify the distinguishing rule. The system uses neural networks to recognize and classify these visual patterns, tackling a benchmark challenge that remains difficult for both humans and algorithms in pattern recognition research.

written from the facts on this page · September 2026

From the sources

Bongard problmes are named after their inventor, Soviet computer scientist Mikhail Bongard, who was working on pattern recognition in the 1960s. He designed 100 of this puzzles, to be a good benchmark for pattern recognition abilities, and they seem to be challenging for both people and algorithms. Here is an example:

Does a similar job

all alternatives →
  • VT
    Visualizing the math that powers 3D character animation2022 · diegomacario.github.io · ▲238

    Hi everyone! I'm the author of this project. I wrote it because I think that the math that makes characters move in games and movies is incredibly beautiful, and I wanted to give others a glimpse into it. It's crazy to think that quaternions, an abstract mathematical tool discovered by William Rowan Hamilton in 1843, would be so perfectly suited to solve hard problems in the world of 3D character animation more than a hundred years later. The story of how he discovered quaternions is also beautiful. Here's an excerpt from Wikipedia: "Hamilton was looking for ways of extending complex numbers…

  • DeepFace – A lightweight deep face recognition library for Python2025 · github.com · ▲257

    DeepFace is a leading open-source library for facial recognition and facial attribute analysis, and the de facto standard in Python. It wraps multiple state-of-the-art models that have reached — and even surpassed — human-level accuracy in recognizing faces. By the numbers (as of early 2025): 15,000+ stars on GitHub; ~4 million installations via pip; 800+ citations in academic papers Whether you're building a cutting-edge AI project or simply exploring facial recognition, DeepFace makes advanced capabilities accessible with just a few lines of code.

  • Visual Search using features extracted from Tensor Flow inception model2016 · github.com · ▲75
  • Quick Draw by Google2016 · quickdraw.withgoogle.com · ▲197

    Can a neural network learn to recognize doodles?

  • Deep Learning for Program Synthesis2017 · microsoft.com · ▲93

    By Rishabh Singh, Jacob Devlin, Abdelrahman Mohamed, and Pushmeet Kohli, Microsoft Research Despite the many advances in computing over the past decades, the actual process of writing computer software has not fundamentally changed — a programmer must manually code the exact algorithmic logic of a p

  • Neural Networks and Deep Learning2015 · neuralnetworksanddeeplearning.com · ▲95

    Core concepts behind neural networks and deep learning

More life & fun this month

the category →

Source: Show HN post ↗

Launched alongside, March 2018

the whole month →
  • Notion 2.0▲2,156

    The all-in-one workspace - notes, tasks, wikis, & databases

    Work · 2018 · notion.so

  • MAKE Book: The Indie Maker Handbook▲1,372

    Learn to build profitable startups the indie way

    Dev tools · 2018 · readmake.com

  • Overflow▲1,139

    The world’s first user flow diagramming tool for designers

    Work · 2018 · overflow.io

  • RoughJS – Create hand-drawn graphics using JavaScript▲1,087

    Rough.js is a light weight javascript library that lets you draw graphics with a hand-drawn, sketchy, appearance.

    Life & fun · 2018 · roughjs.com

  • Sunsama▲999

    If Trello & Google Calendar had a baby.

    Work · 2018 · sunsama.com

  • Airtable Blocks▲877

    500,000+ brands use Airtable to enable real-time collaboration, automate repetitive tasks & manual work, and streamline business processes in minutes. Join them.

    Dev tools · 2018 · airtable.com