Collinearlity, degrees of freedom and LSTM networks

There are many code samples on using LSTM networks for time series prediction. Rather than writing your own LSTM modules and gradient descent computations, usually we will use the off-the-shelf implementations from PyTorch or Tensorflow. Let’s look at a PyTorch model, from data preparation to evaluation: [more]

PXE boot Debian

Diskless system is an old idea (remember Sun Ray 1?) but surprisingly coming back (Raspberry Pi cluster!). I have a system with PXE boot but no disk so it is perfect to try out a diskless system. [more]

DD-WRT on Linksys

I have been using a Linksys WRT3200ACM (with DD-WRT installed) for a while now and I decided to make a better use of such a high-end WiFi router. These are some notes that worth to write down: [more]

Kelly (1956) A new interpretation of information rate

Shannon defined the information rate as the number of bits transmitted per second, which bit is a unit for entropy, i.e., probability is involved. We can relate the information rate with \(R=rH\), where \(r\) is the rate that message is delivered and \(H\) is the entropy of each message. So... [more]

Chatterjee (2015) Modeling credit risk

This is a concise paper describing the Vasicek model. It serves the purpose of describing how banks should evaluate risk on loan assets, and provides a connection between the Black-Scholes option pricing model to pricing bonds or loans. [more]