Toward a Unified Dog Theory: Understanding Pattern Recognition

We might be close in age. I started attending the University of Texas at Arlington (Go Longhorns!) in 1982 with EE as a major. But since I was financing my own college, I ran out of money. And my family, which had promised to help withdrew that help when it was time to back up their words. So, I've never finished a degree.

As for electrical work, I have done everything from digging the ditch to supervising 20 people on a 104,000 square foot school. In fact, I started that school, doing the underground piping. Projects I was fully in charge of ALWAYS finished by the due date, usually with half the crew I was supposed to have, in spite of calendar delays (5 months in one case) and either hit the profit margin or exceeded it. Because my ethic is accomplish the objective regardless of obstacles. And generally, by rewarding crew members of thinking actively and ignoring or redirecting wastes of energy or thought on arguments. I am a money-maker when it comes to running a job. I view the failures of the crew, if any, as failures on my part. My job was to lead them correctly. But I was naive.

I thought the point of being in business was to make money. I liken it to one of my screwdrivers. Every time I pull it out to use, it makes money because it accomplishes some work. So, let me throw that screwdriver away. Well, that is what happened with me. When business got tight, one company decided to keep others not as efficient as I am and let me go. What was the difference between me and the other guys? I haven't known that employer for 25 years like they have. That is, in the end, it was not about making money, it was about who the employer was comfortable being around. And that employer is a really smart guy and I learned a lot from him. Yet, there it is. I was more technically savvy than the job boss on the last project I was on. I could do more than he could do. I have a bigger license. And my jobs didn't run over projected time, like his did. In fact, the project manager started the last big project because the superintendent was still "finishing" his last one.

I stink at politics. So, when people ask me why I can't get hired, I tell them the truth. I have a big license. 25 years experience in the field from basic apprentice to management. An effective way to lead crews to greater efficiency. I make money every time at bat. And I am willing to do any job on the project, without complaint. I also have the trained ability to keep my mouth shut and not speak bad about others but actively look to solve problems caused by oversights. Everything you need to be in business to make money.

So, the question should really be directed at the companies that don't hire me.

I have interviewed for jobs where I have exactly what they are looking for. And asking for 3 dollars an hour less than what I used to make, almost down to apprentice pay. And still not get the job. That leaves the politics or personal feelings of whoever I am dealing with. And I can't help that. What are personal feelings and politics doing in business? I have no idea and it wasn't my idea for that to happen.

Of course, it's not fair. Nor does it help me to be bitter about it. All I can do is hope that others who abuse this process stew in their failings and it causes them to lose sleep.

 My birth place is New Zealand so i need to try and make it make sense to an Amercian. I finished high school in 1976 so i am older than you. My parents made the disconnect then. I did trade/technician quallies, but if i hadn't been so interested in consuming so much alcohol and using various illicit substances i could have got through an EE then. Our country had unbelivably liberal policies for University at that stage. If you passed the entrance exams and/ or were old enough you could go and the fees were tiny. On top of that they paid you quite a nice living allowance to. I then made every lifestyle mistake known to mankind. Early marriage that kind of thing.

In my mid twenties i did a four year degree in maths and stats while teaching. It was the nearest i could get to doing an EE without going full time. In NZ if you had trade quallies and could do the work and done soemthing like that then you were in. When i came here 15 years ago ,they were keen on formal ee quals. So i turned around and did another four year degree distance ed and spread over 8 years. Again if you get in Australa looks after you. The fees are reasonable and you can loan the money off the government if you can't afford it.

Well how did that affect my training? I kept up and if anything extended my dog training to motivate me and keep me going through the second degree.Counter inutitve but it got me through. It sure puts pressure on you to be eifficient in your study and training habits.  I sure have spent more than some time studying how pattern regonition is done. I use neural networks in day ro day work (in a replacement for MR16 quatz halogen down lights believe it or not.!!)

So lets put a strange theory out there.  Lets pretend that for one instant that  machines may well evolve as does nature and that selection pressures casue a kind of evolution to occur. Survival of the fittest shall we say. Then if it is that efficiency after evolution in machines may well give similar solutions to nature then we would say the following

Single pattern dection is efficent and quick and easy

Mulitple patterns take longer computational power.

The more distinct features that we look for the greater the likelhood of a mistake.

Predetermined patterns are much easier to install and organise.

 

So is it patterns that we learn from or is it a behavourist sequnece or or none of the above?

It may well be that pattern recognition  is a skill that many animals have BUT it is most likely to be predetermined without behavourist intervention.It seems like many other patterns are taught often using reinforcement. IT certainly is that scent pattern recognition is very easily taught by behavourist means. Scent to dogs is a very important part of their world vista.

 

 

 

poodleOwned

So is it patterns that we learn from or is it a behavourist sequnece or or none of the above?

It may well be that pattern recognition  is a skill that many animals have BUT it is most likely to be predetermined without behavourist intervention.It seems like many other patterns are taught often using reinforcement. IT certainly is that scent pattern recognition is very easily taught by behavourist means. Scent to dogs is a very important part of their world vista.

 

I'd say it's a little of both. This sounds like comparing Classical Conditioning with Operant Conditioning. I think of Classical Conditioning as being associative learning and Operant Conditioning as learning in sequence. With associative learning there seems to be a wide range with "what" animals associate with "what". That "what" could also include "who", "when", "where", and then incorporates OC into the "how" something is remembered and acted upon in the future. In marker training I see a pairing of CC and OC.

There may also be something to be learned from the "fast mapping" studies:

https://webspace.utexas.edu/bighamds/LIN312/Files/dog-vocab.pdf

As far as scent goes, I understand that scent is the one of the five senses most connected with memory. I've often wondered if due to the dog's ultra keen sense of smell, they have a more heightened ability to make connections and remember certain events through scent imagery and the feelings or frame of mind they were in when a scent from their past is presented. I know that even as a human with nothing like the canine nose going for me, a scent can bring back powerful images and feelings of places, events, and other living beings from my past. And, it works the other way around (at least for me) in that if I think back on a place or event, the memory of what these things smelled like is quite vivid and enhances the memory.

poodleOwned

 My birth place is New Zealand so i need to try and make it make sense to an Amercian. I finished high school in 1976 so i am older than you. My parents made the disconnect then. I did trade/technician quallies, but if i hadn't been so interested in consuming so much alcohol and using various illicit substances i could have got through an EE then. Our country had unbelivably liberal policies for University at that stage. If you passed the entrance exams and/ or were old enough you could go and the fees were tiny.

So lets put a strange theory out there.  Lets pretend that for one instant that  machines may well evolve as does nature and that selection pressures casue a kind of evolution to occur. Survival of the fittest shall we say. Then if it is that efficiency after evolution in machines may well give similar solutions to nature then we would say the following

Single pattern dection is efficent and quick and easy

Mulitple patterns take longer computational power.

The more distinct features that we look for the greater the likelhood of a mistake.

Predetermined patterns are much easier to install and organise.

 

So is it patterns that we learn from or is it a behavourist sequnece or or none of the above?

It may well be that pattern recognition  is a skill that many animals have BUT it is most likely to be predetermined without behavourist intervention.It seems like many other patterns are taught often using reinforcement. IT certainly is that scent pattern recognition is very easily taught by behavourist means. Scent to dogs is a very important part of their world vista.

 

A kiwi, eh? I've known a few kiwis here in Texas. I also happen to know that the nickname comes from the bird, now extinct, not the fruit. I also tried to find out if there is a dinstinction drawn between tongans and maoris, something I read about in a fictional book. In the book, "Friday," by Heinlein, the main character has a home for a while in Christchurch. Anyway, then I got to see it, I think, or at least a representation of it in the movie "8 Below." Walker's character is trying to get back to Antarctica to rescue his dogs and Christchurch is the southern most port o' call to get there, at least by the route he was taking. And I totally dig the NZ accent. It has a sing-song quality. I knew a kiwi in college and she had the most delightful quality in her voice. She even liked me, somewhat. Enough trivia and rambling.

I agree with your definitions and conditions of pattern recognition. I also don't see it as separate from behavioral sequence. Dogs, like any successful organism, are highly efficient in terms of energy use and reflex. And I think that the definition crosses species. We learn simple things quicker. And it is across species, too. And using treats for training is using that pattern recognition. Sitting when a human says "Sit" equals reward. "Down" when a human says that word equals treat. Eventually, listening to humans equals good things. I think this happens in parallel to what dogs seek on their own.

So the dog digs in the garden. Instinctually (for no instantly visible reason) or for a purpose, such as to get a worm. The dog will also listen to you for other things but he will also dig when it pleases him. Unless you can reward him for not digging. And that only works if he has a greater desire for your reward than the digging. Likewise with correction. Avoiding the correction, even if he sees it as a correction, may not be enough to stop digging. Both of these reasonings are sure to offend some because it means the dog "has a mind of his own." It also is certain to tick of some because it tends to support the notion that dogs are NOT mindless automatons. Or, the dog will quit digging while he is paying attention to you and later, will go back to it. Or, he may quit digging after one or two interventions, correction, reward, or combo.

To me, treat training is about as natural and simple as it gets. Ron = good things. Ron = survival. I haven't subverted nature, I am working with nature.

And smell is important, yes. My smoked brisket is worth a couple of doggy sonnets, at least. Food .... eat ... survive. How do I get that food? Following the path of least resistance, as all creatures do, do what Ron says.

Angelique, I used to think like you that the initial introduction of clicker (called charging, where the dog links clicker with reward, usually a food treat, but it can be anything) was associative. Now, I think it is still OC. The dog learns that click equals food (or whatever reward). And perhaps I have used associative, i.e., classical, in the past because I had not looked deep enough to see what the reward or even avoidance of punishment was. And even the non-linear dog theory of maintaining social and emotional equilibrium has, in itself, the process of reward. Even if we wanted to decide that equilbrium was not either inherently or overtly rewarding, it is assumed in the theory to be a desired state in dogs or other creatures. Desired over imbalance, specifically. Reward, release, equilibrium, potayto, potahto.

So, IMHO, calling something classical or associative because it lacked the overt presentation of a reward by a human does not negate that a reward sequence was not involved and, ergo, does not prove that OC was not happening.

ron2

Angelique, I used to think like you that the initial introduction of clicker (called charging, where the dog links clicker with reward, usually a food treat, but it can be anything) was associative. Now, I think it is still OC. The dog learns that click equals food (or whatever reward). And perhaps I have used associative, i.e., classical, in the past because I had not looked deep enough to see what the reward or even avoidance of punishment was.

 

This is what I'm referring to:

 

"Secondary Positive Reinforcers and Bridges"

"A primary positive reinforcer is something that the animal does not have to learn to like. It comes naturally, no experience necessary. Primary R+s usually include food, often include sex (the chance to mate), the chance to engage in instinctive behaviors, and for social animals, the chance to interact with others.

A secondary positive reinforcer is something that the animal has to learn to like. The learning can be accomplished through Classical Conditioning or through some other method. A paycheck is a secondary reinforcer - just try writing a check to reward a young child for potty training!

Animal trainers will often create a special secondary reinforcer they call a bridge. A bridge is a stimulus that has been associated with a primary reinforcer through classical conditioning. This process creates a conditioned positive reinforcer, often called a conditioned reinforcer or CR for short. Animals that have learned a bridge react to it almost as they would to the reward that follows (animals that have learned what clicker training is all about may sometimes prefer the CR that tells them they got it right to the actual "reward";)."

Thanks for the clarification.

 

ron2

A kiwi, eh? I've known a few kiwis here in Texas. I also happen to know that the nickname comes from the bird, now extinct, not the fruit. I also tried to find out if there is a dinstinction drawn between tongans and maoris, something I read about in a fictional book. In the book, "Friday," by Heinlein, the main character has a home for a while in Christchurch. Anyway, then I got to see it, I think, or at least a representation of it in the movie "8 Below." Walker's character is trying to get back to Antarctica to rescue his dogs and Christchurch is the southern most port o' call to get there, at least by the route he was taking. And I totally dig the NZ accent. It has a sing-song quality. I knew a kiwi in college and she had the most delightful quality in her voice. She even liked me, somewhat. Enough trivia and rambling.

 

 

I still have a full on kiwi accent. The kiwi is still around just not very many of them. Kiwifruit are actually chinese gooseberries and is one of the best marketing jobs ever foisted on mankind, truely awesome. The distinction between tongans and maoris is where they settled, the language is remarkably similar. I grew up in Christchurch which is just had a terrible terrible earthquake. We are a remarkable people, tiny country (4 million people). !0% of our population lives away from NZ.  We have the world's best Rugby Union Team the All Blacks of which i am insanely proud of. We also lifted the America's cup of the US (Yachting Trophy).  Enough trivia. Oh and we are just the world's best dog handlers! :)

poodleOwned

I still have a full on kiwi accent. The kiwi is still around just not very many of them. Kiwifruit are actually chinese gooseberries and is one of the best marketing jobs ever foisted on mankind, truely awesome. The distinction between tongans and maoris is where they settled, the language is remarkably similar. I grew up in Christchurch which is just had a terrible terrible earthquake. We are a remarkable people, tiny country (4 million people). !0% of our population lives away from NZ.  We have the world's best Rugby Union Team the All Blacks of which i am insanely proud of. We also lifted the America's cup of the US (Yachting Trophy).  Enough trivia. Oh and we are just the world's best dog handlers! :)

And according to Friday's description in the book all of NZ is paradise, though the character had a special liking for Christchurch. The gardens. The center of town. Beautiful architecture. And good people, white, tongan, or maori.

New research shows that microglial cells, whose only function was thought to be to fight off infections in the brain, are now shown to play an important role in detecting environmental changes.

LCK

ron2

Angelique, I used to think like you that the initial introduction of clicker (called charging, where the dog links clicker with reward, usually a food treat, but it can be anything) was associative. Now, I think it is still OC. The dog learns that click equals food (or whatever reward). And perhaps I have used associative, i.e., classical, in the past because I had not looked deep enough to see what the reward or even avoidance of punishment was. And even the non-linear dog theory of maintaining social and emotional equilibrium has, in itself, the process of reward. Even if we wanted to decide that equilbrium was not either inherently or overtly rewarding, it is assumed in the theory to be a desired state in dogs or other creatures. Desired over imbalance, specifically. Reward, release, equilibrium, potayto, potahto.

So, IMHO, calling something classical or associative because it lacked the overt presentation of a reward by a human does not negate that a reward sequence was not involved and, ergo, does not prove that OC was not happening.

 

Interesting take: http://brembs.net/learning/drosophila/general_introduction.html