Google Intern vs Media AI

Google Intern : Media : from keras.layers import * from keras.models import * from .data import load_data x, y, x_test, y_test = load_data() def get_model(num_layers): model = Sequential() for _ in range(num_layers): model.add(Dense(100, activation='sigmoid')) model.compile(loss='mse', optimizer='sgd') return model best_model = None best_loss = None for i in range(1, 10): model = get_model(i) model.fit(x, y) loss = model.evaluate(x_test, y_test) if best_loss is None or loss < best_loss: best_loss = loss best_model = model This Google AI created its own "baby" AI

A split-screen meme contrasts a Google intern writing a simple neural-network training script with the media presenting a humanoid robot as an AI that created its own “baby” AI. The joke mocks sensationalist media coverage and the exaggeration of ordinary machine-learning work into a dramatic breakthrough. EN: A split-screen meme contrasts a Google intern writing a simple neural-network training script with the media presenting a humanoid robot as an AI that created its own “baby” AI. The joke mocks sensationalist media coverage and the exaggeration of ordinary machine-learning work into a dramatic breakthrough.

• @profunctor_io / 572 2018-01-20