Applied learning optimisation

Imagine a particular posterior distribution is represented as a two dimensional surface with ridges and contours. During training different aspects of this distribution become apparent to the agent. Say at time t=n the amount of the distribution revealed is S(n). we could imagine a small change in learning to be dS/dt to reveal a bit more information about the distribution. Now in order for something to exist , wether information or matter, it requires a form..and form is a function of contrast. Or rather A is = A because A!= !A (the axiom of identity. a thing is what it is because it can be contrasted with what its not. So any feature of the posterior distribution must exist because it contrasts with another part of the posterior distribution. So if we are developing a ridge there must be parts that are lower than others and parts that are higher than others. When we observe dS/dt we reveal a little more of where this change is headed...which means we can optimise the learning prosses by looking towards the limit of this change. The power with this is not contained in the training of an original agent. once we have observed the difference between The actual dS/dt and the predicted dS/dt the agent will get a feel of how such changes naturally follow in order to be a more succesful predictor. What it means is that learning to classify cats and non cats will have the same class of posterior distributions as trucks and non trucks, and we are on our way transfering knowledge from one to the other. what we are transfering is the nature of the progression taken to build a posterior distribution from one class to another.

Comments

Post a Comment

Popular posts from this blog

ABSTRACT submited

BACKPROPAGATION ON A NOVEL NETWORK (NON ANN) FOR HATESPEECH DETECTION AND MORE This paper introduces a novel type of network that can accurately classify "hate speech" and differentiate it from normal content. It develops a network centered around a unique POS (part of speech) tagging type of system whose specifics are learnt by the network. Each word is assigned variables that indicate the polarity, (sign) of the effect a particular word has on every other word in a sentence, the magnitude of that polarity and also to what extent the word filters itself from the effect from other words. When choosing a set of properties, we would like to properly understand what we want the network to do. We know that both posts classified as hate speech and those not contain samples of words from a common pool (they share words). But we still want that the results for opposite classes to polarize. On top of this I believe that this represents a complex system, in that small chang...
Read more
This involves a "hate speech" or "spam" detection algorithm. (depending on the training data). This will be based on ANN theory , and involve the backpropagation algorithm while not based on an actual neural network. The training data will be labeled into two classes, "hate speech" and "non hate speech".   Hate speech will have an expected value of 1 , representing certain hate speech, while non hate speech will have an expected value of -1.   So what we would like is for the software to be input a post and gives it a value of 1 or -1, or something much closer to one or the other. During training The program will compute a value by performing the equivalent of forward propagation through the parameters we assign to each of multiple structures we will embed in the input ...then using this value and the epected value we will compute a cost function. We will then backpropagate through the heirachical structure of parameters in the direc...
Read more