Dropout Defender Game

Defend neural networks from overfitting by tuning dropout, watching neurons deactivate, comparing results, completing challenges, and discovering stronger generalisation strategies interactively today visually now.

Game score
0
Run a comparison
Validation score
Dropout model
Generalisation gap
Lower is better
Active neurons
Expected per batch
Best epoch
Validation minimum
Player level
1
0 XP
Neural defence arena Ready
Awaiting training Round 1
0 / 0 epochs
Challenge setup?
Find the Best Rate
Network architecture0 params
0.30
0.30
0.20
Training controls?
600
0.20
70%
0.000
0.000
Training command centreTensorFlow.js
Ready: Configure a challenge, then compare baseline and dropout models.
Experiment results0 experiments
TimeModelDatasetDropoutTrainValidationTestGapEpochsScore
No experiments yet.
Achievements and progress0/6 unlocked

Local leaderboard

#PlayerChallengeScore
Dropout learning guideEducational mode

How dropout works

During training, dropout independently removes selected activations with probability p. Remaining activations are scaled so inference can use every neuron without changing the expected signal.

Generalisation gap

The gap compares training performance with validation performance. A large positive gap often signals memorisation, while weak scores on both sets can indicate underfitting.

Recommended starting ranges

Dense hidden layers often start around 0.10 to 0.50. Input dropout is usually lower. The best value depends on data size, noise, network capacity, and regularisation.

Other defences

L1 promotes sparse weights, L2 discourages large weights, and early stopping limits memorisation time. Excessive combined regularisation may underfit the model.

Formula and terminology

Mask: mᵢ ~ Bernoulli(1 − p). Training activation: h̃ᵢ = mᵢhᵢ/(1 − p). Expected active neurons: n(1 − p).

Co-adaptation: neurons depending too heavily on specific partners. Inverted dropout: scaling during training instead of inference.

Related Calculators

Build a Neural NetworkNeuron Activation GameActivation Function MatchBackpropagation PuzzleWeight Adjustment ChallengeDense Layer Output GameVanishing Gradient EscapeNeural Network Architecture BuilderLoss Function Challenge

Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.