It's challenging to maintain the balance in life.

Sometimes hard to reach.

Time has been past.

I feel like I can't keep up with life sometimes. Since last week, my sister has had a rash, and we spent too much time in the hospital. We went to the emergency room at night and in the mornings. I was planning to prepare my thesis presentation; however, life has just stopped, and my... Continue Reading →

In classification, every prediction eventually enters one of four boxes: true positive, false positive, false negative, or true negative. Regression is less tidy. A house-price prediction can be wrong by $1,000, $10,000, or $100,000. All three are errors, but they do not tell the same story. For this final post, we will use the publicly... Continue Reading →

Suppose a bank wants to predict which customers are likely to subscribe to a term deposit after a marketing campaign. For this example, we will use the publicly available UCI Bank Marketing dataset. The complete bank-full.csv file contains 45,211 customer records: 39,922 did not subscribe 5,289 subscribed Only about 11.7% belong to the positive class.... Continue Reading →

Mini-batch gradient descent does not follow a perfectly smooth path. One batch pushes the model slightly left. The next pushes it right. In a narrow valley, the optimizer may spend much of its energy bouncing between the walls instead of moving toward the bottom. Plain gradient descent reacts only to the current gradient. Momentum, RMSProp,... Continue Reading →

Gradient descent tells us to calculate the gradient and update the model’s parameters. There is one detail hidden inside that sentence: How much data should we use to calculate each gradient? We could use the entire training set. We could use one example. Or we could choose a small group somewhere between the two. These... Continue Reading →

We have chosen a search method. We have considered initialization, architecture, batch size, learning rate, and regularization. Now comes the part that looks deceptively simple: Run the tuner and keep the winner. Unfortunately, a good hyperparameter-tuning workflow contains several boundaries. If we cross them carelessly, we can produce a model with an excellent score and... Continue Reading →

Imagine walking downhill in thick fog. You cannot see the bottom, so you estimate the slope beneath your feet and take a step. Three decisions matter: How much ground do you inspect before choosing a direction? How large a step do you take? How strongly do you avoid complicated or risky paths? In neural-network training,... Continue Reading →