🔬 Research
Distill.pub
25 min read
Adversarial Reprogramming of Neural Cellular Automata
Reprogramming Neural CA to exhibit novel behaviour, using adversarial attacks.
Research
75 articles
🔬 Research
Distill.pub
11 min read
Weight Banding
Weights in the final layer of common visual models appear as horizontal bands. We investigate how and why.
🔬 Research
Distill.pub
13 min read
Branch Specialization
When a neural network layer is divided into multiple branches, neurons self-organize into coherent groupings.
🔬 Research
Distill.pub
9 min read
Multimodal Neurons in Artificial Neural Networks
We report the existence of multimodal neurons in artificial neural networks, similar to those found in the human brain.
🔬 Research
Distill.pub
30 min read
Self-Organising Textures
Neural Cellular Automata learn to generate textures, exhibiting surprising properties.
🔬 Research
Distill.pub
15 min read
Visualizing Weights
We present techniques for visualizing, contextualizing, and understanding neural network weights.
🔬 Research
Distill.pub
3 min read
Curve Circuits
Reverse engineering the curve detection algorithm from InceptionV1 and reimplementing it from scratch.
🔬 Research
Distill.pub
18 min read
High-Low Frequency Detectors
A family of early-vision neurons reacting to directional transitions from high to low spatial frequency.
🔬 Research
Distill.pub
18 min read
Naturally Occurring Equivariance in Neural Networks
Neural networks naturally learn many transformed copies of the same feature, connected by symmetric weights.
🔬 Research
Distill.pub
40 min read
Understanding RL Vision
With diverse environments, we can analyze, diagnose and edit deep reinforcement learning models using attribution.
🔬 Research
Distill.pub
36 min read
Communicating with Interactive Articles
Examining the design of interactive articles by synthesizing theory from disciplines such as education, journalism, and visualization.
🔬 Research
Distill.pub
5 min read
Thread: Differentiable Self-organizing Systems
A collection of articles and comments with the goal of understanding how to design robust and general purpose self-organizing systems.
🔬 Research
Distill.pub
27 min read
Self-classifying MNIST Digits
Training an end-to-end differentiable, self-organising cellular automata for classifying MNIST digits.
🔬 Research
Distill.pub
36 min read
Curve Detectors
Part one of a three part deep dive into the curve neuron family.
🔬 Research
Distill.pub
29 min read
Exploring Bayesian Optimization
How to tune hyperparameters for your machine learning model using Bayesian optimization.
🔬 Research
Distill.pub
17 min read
An Overview of Early Vision in InceptionV1
An overview of all the neurons in the first five layers of InceptionV1, organized into a taxonomy of 'neuron groups.'
🔬 Research
Distill.pub
42 min read
Visualizing Neural Networks with the Grand Tour
By focusing on linear dimensionality reduction, we show how to visualize many dynamic phenomena in neural networks.
🔬 Research
Distill.pub
5 min read
Thread: Circuits
What can we learn if we invest heavily in reverse engineering a single neural network?
🔬 Research
Distill.pub
42 min read
Zoom In: An Introduction to Circuits
By studying the connections between neurons, we can find meaningful algorithms in the weights of neural networks.
🔬 Research
Distill.pub
24 min read
Growing Neural Cellular Automata
Training an end-to-end differentiable, self-organising cellular automata model of morphogenesis, able to both grow and regenerate specific patterns.
🔬 Research
Distill.pub
34 min read
Visualizing the Impact of Feature Attribution Baselines
Exploring the baseline input hyperparameter, and how it impacts interpretations of neural network behavior.
🔬 Research
Distill.pub
27 min read
Computing Receptive Fields of Convolutional Neural Networks
Detailed derivations and open-source code to analyze the receptive fields of convnets.
🔬 Research
Distill.pub
17 min read
The Paths Perspective on Value Learning
A closer look at how Temporal Difference Learning merges paths of experience for greater statistical efficiency
🔬 Research
Distill.pub
9 min read
A Discussion of 'Adversarial Examples Are Not Bugs, They Are Features'
Six comments from the community and responses from the original authors