Reinforcement Learning

In computer science, the term reinforcement learning refers to a sub-study of Artificial Intelligence called machine learning. The subject itself is actually quite broad and deep, and propels you quickly into the world of probabilities, Markov Decision Processes, weights, and rewards. Thus it is certainly not a subject I intend to discuss in detail here.

But outside of the field of computer science, this term is used and should already be understood by the majority of you. Be it by your masterful comprehension of English, or perhaps by your familiarity with the story of Pavlov’s dog.

The concept itself is quite simple really, and simply proves that if you repeatedly derive a “good” or “bad” outcome from a meaningless action, that action will gradually start to be considered “good” or “bad” in and of itself. In the case of Pavlov’s dog, this meant that ringing the bell eventually became associated with “food” in the dog’s canine brain, and soon that dog began to salivate at the mere sound of the bell (whether or not food followed close behind). In the field of Psychology, this is known as a conditioned response.

So what does all that have to do with me today? Well today for the first time in many months, I had to take a long walk outside. About 40 blocks, I think. Still making no sense? Then let me break it down for you.

You see when I stepped outside today, it was a nice warm day, the sun was shining, the birds were chirping, and well, it suffices to say that spring has come in full force to Philadelphia. This instantly put me in a great mood, and I began to walk briskly to my destination. It was a nice day, I had things to do and places to be, and moreover, I wanted to get those things over with so I could make love to the sun. 😆

It was a nice day, I had things to do and places to be, and moreover, I wanted to get those things over with so I could make love to the sun.

Then I started to feel the aches of anaerobic exercise. I could literally feel the lactic acid accumulating in my thighs, calves, and feet. I wasn’t running, I wasn’t jogging, I was merely walking as I pleased. Well, I was, but now I’m limping. The journey so far? Just 3 blocks.

So I slowed down before my legs gave way. But it was still a nice day, I still had places to be, and I still wanted time to make my date with the sun. So this didn’t last long either. I unconsciously picked up my pace again, and again the muscle fatigue began to set in. This time I found my breathing was irregular. I wasn’t quite out of breath, but I wasn’t in control of it either. The journey so far? Now on block 7.

Thus began the cycle of speed up, slow down, speed up by a little less, slow down, and so on and so forth ad nauseum. We are now on block 12.

By block 15, I was walking without a limp, without losing my breadth, and just kind of taking my time to enjoy the sun’s presence. At this speed, I might as well give her the bouquet, and set the table now. You know what they say, there is no time like the present.

So I made it to block 20 whilst taking a leisurely stroll in the sunlight, and to my surprise I had actually made acceptable time. The trip was about a 40 minute walk (where I had predicted 30 from previous experience). My walk so far had been fraught with fatigue, stumbling, and shortness of breadth.

Consequently, it was only on my return trip, that I truly began to appreciate my acquired slow pace. At this speed, not only could I enjoy my walk in the sun and let my thoughts wander, but I could now also afford a brief jog/run when necessary: like in an emergency, or more commonly, a changing traffic light. 😆

So I lit the candle and popped the champagne on the way back. The sun and I had a nice chat, and we even got a little hot and sweaty in a public area.

My 40 block journey had been most enjoyable! But more so than that, it had been revealing. In the time spent letting my thoughts wander, I had come to realize that my walk, was the perfect metaphor for the effects of living with sickle cell anaemia!

Our bodies force a conditioned response onto us, and though forced, we learn to cope and appreciate our bodies for the benefits of behaving that way with them. This might explain why like Pavlov’s dog, all sickler’s feel an attachment to their disease. Somewhere in our primitive brains, the connection gets made between the meaningless substance of behaving like a sickler, and the rewards of behaving like a sickler in a sickler’s body.

It certainly provides a theory for why when I first heard about my friend Bisade’s recent foray into getting cured of SCD, I was very conflicted about attempting the same. Why should I do this, when I already know how to work my current body? Seemed like a whole lot of trouble, for a chance to maybe not reject the transplant in the end… and for a body that I don’t even know how to use should I get it.

But if we can all admit that Pavlov’s experiment simply went to show how primitive canine brains are, and how easily they can be fooled. Then by extension, we can induce that this line of reasoning is conditioned in our own primitive brains, and thus probably not the smart response.

So if you are a sickler like Bisade and I, stop by his blog, see what the big deal is about, and then make an informed decision. There is no time like the present!

There is no time like the present!

Damola Mabogunje

Software Engineer by day, Blogger, Poet and Author, by night, I spend my days writing everything from the convex comma, to highly complex code.

One thought on “Reinforcement Learning

Leave a Reply

Your email address will not be published. Required fields are marked *