I recently started playing with neural nets via TensorFlow. It’s a big departure to my reinforcement learning background. But I must say, the learning is usually the other way around for most people. That is, they first learn neural nets, then reinforcement learning. That being said, many concepts overlap. One of the many things that… Continue reading Handwritten Digit Recognition with TensorFlow
Category: Artificial Intelligence
Tile Code – Generalization
Coarse Coding, such as Tile Coding, is a nice way of storing a model of the environment linear in nature or have ordinal characteristics (for instance, alphabets are ordinal since it can be ordered: A < B < C, …). The image above illustrates how Coarse Coding may be utilized. a) Narrow Generalization is for… Continue reading Tile Code – Generalization
SDM And Database Brainstorm
With my Sparse Distributed Memory (SDM) basic implementation finally done, in which the Up/Down Counters won’t overflow, I’m ready to embark something bigger. At the moment, my implementation stores everything in random-access-memory (RAM), which is very limiting. I have utilized SDM with my reinforcement learning (rl), solving entailments in a set of 11,000 metrics, which… Continue reading SDM And Database Brainstorm
Sparse Distributed Memory – Addressing the Up/Down Counter Overflow
Sparse Distributed Memory (SDM) is a fascinating way of managing the exponential growth in memory. As opposed to the traditional random-access memory (RAM) where there is a one to one correspondence between an address and a hard location, SDM can map multiple addresses to a multiple of hard location. I won’t delve into why this… Continue reading Sparse Distributed Memory – Addressing the Up/Down Counter Overflow