Machine Learning : Intro

 

Introduction

  Machine learning, artificial intelligence (AI), and cognitive computing are dominating conversations about how emerging advanced analytics can provide businesses with a competitive advantage to the business. There is no debate that existing business leaders are facing new and unanticipated competitors. These businesses are looking at new strategies that can prepare them for the future. While a business can try different strategies, they all come back to a fundamental truth — you have to follow the data. 

What Is Learning?

  Let us begin by considering a couple of examples from naturally occurring animal learning. Some of the most fundamental issues in ML arise already in that context, which we are all familiar with.

Bait Shyness – Rats Learning to Avoid Poisonous Baits: When rats encounter food items with novel look or smell, they will first eat very small amounts, and subsequent feeding will depend on the flavor of the food and its physiological effect. If the food produces an ill effect, the novel food will often be associated with the illness, and subsequently, the rats will not eat it. Clearly, there is a learning mechanism in play here – the animal used past experience with some food to acquire expertise in detecting the safety of this food. If past experience with the food was negatively labeled, the animal predicts that it will also have  anegative effect when encountered in the future.

  Inspired by the preceding example of successful learning, let us demonstrate a typical machine learning task. Suppose we would like to program a machine that learns how to filter spam e-mails. A naive solution would be seemingly similar to the way rats learn how to avoid poisonous baits. The machine will simply memorize all previous e-mails that had been labeled as spam e-mails by the human user. When a new e-mail arrives, the machine will search for it in the set of previous spam e-mails. If it matches one of them, it will be trashed. Otherwise, it will be moved to the user’s inbox folder.

  While the preceding “learning by memorization” approach is sometimes use-

ful, it lacks an important aspect of learning systems – the ability to label unseen e-mail messages. A successful learner should be able to progress from individual examples to broader generalization. This is also referred to as inductive reasoning or inductive inference. In the bait shyness example presented previously, after the rats encounter an example of a certain type of food, they apply their attitude toward it on new, unseen examples of food of similar smell and taste. To achieve generalization in the spam filtering task, the learner can scan the previously seen e-mails, and extract a set of words whose appearance in an e-mail message is indicative of spam. Then, when a new e-mail arrives, the machine can check whether one of the suspicious words appears in it, and predict its label accordingly. Such a system would potentially be able correctly to predict the label of unseen e-mails.

 However  inductive reasoning might lead us to false conclusions. To illustrate

this, let us consider again an example from animal learning.

Pigeon Superstition: In an experiment performed by the psychologist B. F. Skinner,

he placed a bunch of hungry pigeons in a cage. An automatic mechanism had been attached to the cage, delivering food to the pigeons at regular interval with no reference whatsoever to the birds’ behavior. The hungry pigeons went around the cage, and when food was first delivered, it found each pigeon engaged in some activity (pecking, turning the head, etc.). The arrival of food reinforced each bird’s specific action, and, consequently each bird tended to spend some more time doing that very same action. That, in turn, increased the chance that the next random food delivery would find each bird engaged in that activity. again What results is a chain of events that reinforces the pigeons’ association of the delivery of the food with whatever chance actions they had been performing when it was first delivered. They subsequently continue to perform these same actions diligently.

 What distinguishes learning mechanisms that result in superstition from useful learning? This question is crucial to the development of automated learners.

 While human learners can rely on common sense to filter out random meaningless learning conclusions, once we export the task of learning to a machine, we must provide well defined crisp principles that will protect the program from reaching senseless or useless conclusions. The development of such principles is a central

goal of the theory of machine learning.

What, then, made the rats’ learning more successful than that of the pigeons?

 As a first step toward answering this question, let us have a closer look at the bait shyness phenomenon in rats.

Bait Shyness revisited – rats fail to acquire conditioning between food and electric shock or between sound and nausea: The bait shyness mechanism in rats turns out to be more complex than what one may expect. In experiments carried out by Garcia, it was demonstrated that if the unpleasant stimulus that follows food consumption is replaced by, say, electrical shock (rather than nausea), then no conditioning occurs. Even after repeated trials in which the consumption of some food is followed by the administration of unpleasant electrical shock, the rats do not tend to avoid that food. Similar failure of conditioning occurs when the characteristic of the food that implies nausea(such as taste or smell) is replaced by a vocal signal. The rats seem to have some “built in” prior knowledge telling them that, while temporal correlation between food and nausea can be causal, it is unlikely that there would be a causal relationship between food consumption and electrical shocks or between sounds and nausea.

 We conclude that one distinguishing feature between the bait shyness learning

and the pigeon superstition is the incorporation of prior knowledge that biases the learning mechanism. This is also referred to as inductive bias. The pigeons in

the experiment are willing to adopt any explanation for the occurrence of food.

However, the rats “know” that food cannot cause an electric shock and that the co-occurrence of noise with some food is not likely to affect the nutritional value of that food. The rats’ learning process is biased toward detecting some kind of patterns while ignoring other temporal correlations between events.

  It turns out that the incorporation of prior knowledge, biasing the learning process, is inevitable for the success of learning algorithms. The development of tools for expressing domain expertise, translating it into a learning bias, and quantifying the effect of such a bias on the success of learning is a central theme of the theory of machine learning. Roughly speaking, the stronger the prior knowledge (or prior assumptions) that one starts the learning process with, the easier it is to learn from further examples. However, the stronger these prior assumptions are, the less flexible the learning is – it is bound, a priori, by the commitment to these assumptions. For more you can see the What is Machine Learning? post.