How to keep the kids learning during school closures?

So, the kids are coming home for the foreseeable future. Schools are going to send work home, and we all get to be teachers for the next few months. That said, not all schools can do this, and they won’t necessarily be able to cover all topics.

With all this in mind, it seems like a good idea to produce a list of resources that may be useful over the coming months. I’ll try to update this list as I find more sources. Initially I’ll start with a list of sites I already know of, including some that my kids’ schools use.

https://www.bbc.co.uk/bitesize – UK based – Owned by the BBC.

https://www.mymaths.co.uk/ – UK based – Owned by the Oxford University Press

https://www.rosettastone.co.uk/ – USA based – Commercial online learning provider

https://www.kerboodle.com/users/login – UK based – Owned by the Oxford University Press

https://www.doddlelearn.co.uk/ – UK based – Commercial online learning provider

https://www.vocabexpress.com/co/ – UK based – Owned by Collins

https://www.futurelearn.com/ – UK based – Owned by the Open University

https://www.khanacademy.org/ – USA Based – Non-profit online learning provider

https://online.stanford.edu/courses – USA based – Owned by Stanford University

https://ocw.mit.edu/index.htm – USA based – Owned by Massachusetts Institute of Technology

https://www.edx.org/ – USA based – Owned by Massachusetts Institute of Technology and Harvard University

https://www.codecademy.com/ – USA based – Commercial online learning provider

https://www.p2pu.org/ – USA based – Non-profit online learning provider

https://alison.com/ – Ireland based – Commercial online learning provider

https://www.udacity.com/ – USA based – Commercial online learning provider

https://www.lynda.com/ – USA based – Owned by LinkedIn

https://www.udemy.com/ – USA based – Commercial online learning provider

https://www.thegreatcourses.com/ – USA based – Commercial online learning provider

https://www.coursera.org/ – USA based – Commercial online learning provider

https://developers.google.com/training/ – USA based – Owned by Google

I’ll try to add to this list as times goes on. That said, there are some excellent sites for those seeking access to online learning. One that I’ve used in the past, for example, is https://www.mooc-list.com/. If I find any others, I’ll add them to this post.

https://www.frugalmum.co.uk/free-teaching-resources/ – A resource I found shared on Facebook

Last updated: 20/03/2020

 

How to approach Covid-19?

I find myself in a bizarre position watching the spread of Covid-19. Like many, I find myself worrying about family members. Two of my kids are asthmatic, my Mum has emphysema, I had five pneumothoraxes when I was younger and lung cancer is common in my family. We don’t tend to do well where lungs are concerned.

Leaving that aside, though, I find myself incredibly curious. When I went into my first degree, it was because I was going to be Cracker. I was going to solve crime and save lives with my knowledge of criminal psychology (while smoking like a chimney and drinking whisky by the gallon).

By the time I finished my degree I had discovered evolutionary psychology. I still wanted to be a forensic psychologist, which didn’t happen, but I’d also fallen in love with a combination of statistics, genetics and evolutionary theory. I was especially enamoured of the Red Queen hypothesis, an evolutionary arms race with parasites/illnesses.

Now I’m doing a degree in computing and IT and I’ve rediscovered that love, while completing modules discussing maths and robotics/AI. Alongside that, of course, we’ve had the appearance of Covid-19. This has led me to shift my focus somewhat, in the past week or so, and begin to wonder about how contagions work.

The problem, though, is I’m drowning in expert analysis but not in data. I want to remain calm about Covid-19, but the dearth of UK data and modelling to study is a barrier to that. Obviously there are plenty of people saying the UK strategy is wrong, and I tend to agree, but without the data I’m just following the herd (see what I did there).

As I can’t get a decent look at the UK data and modelling, I’m trying to learn more about the subject generally. I’m reading The Rules of Contagion by Adam Kucharski. I’m taking advantage of access to journals too. Essentially I’m trying to stick to expert information sources, but doing my own research rather than just waiting for information.

Of course none of this makes decision making any easier. Do I “stock up”? Do I take the kids out of school? Should I be seeking to work from home? These questions are all stuck at the back of my mind, elevating my stress levels. But reading, learning and using the right sources are all helping. I therefore recommend that everyone do the same.

 

 

Bee Bot pollinator

A simple idea that came to me in a tutorial today relates to the Black Mirror episode about robotic bees.

The basic premise of robotic bees is that they carry out the role of real bees as pollinators. The bees in Black Mirror seemed to be quite complex, and a bit too realistic. Could we create a system of pollinators that is essentially very simple?

The key components, I think, are around the ability to move, the ability to collect and deposit pollen, the ability to sense pollen, the ability to sense flowers. and the ability to sense objects to avoid collision.

I don’t think we need a bee bot pollinator that is the size of a bee, or one that can fly. The key considerations for me, around size, actually relate to what sensors, actuators and power sources are practical for such a purpose.

In terms of what the bee might do, I suspect that its behaviour could largely be random, within certain constraints. Specifically, once it senses pollen, it should move towards it. Once it has pollen, it should move for a set distance, then it should move towards flowers.

Of course, it may not sense pollen or flowers, so it should then move randomly, probably within a specific area, perhaps avoiding a previous location. This area could be controlled by something as simple as fencing, or a more complex system of radio signals.

This basic system obviously isn’t quite so simple in terms of sensors and programming. It could, however, form the basis of a robot not hugely more complex than the Turtle robot originally created by William Grey Walter. It might make an interesting project anyway.

 

 

Is now the time to move more things online?

I wouldn’t want to be in the airline business, tourism business or some elements of the service sector at the moment. That said, I expect products like Adobe Connect, Zoom, Slack, Discord, and the various services Microsoft and Google offer might be about to come into their own.

From a number of points of view, this could be an extremely positive thing. Firstly, it may help prevent the spread of Covid-19. It may also help with future outbreaks of illness. Secondly, it will probably save some businesses a lot of money. This will ultimately depend on the relative expense of flying and upgrading tech.

A third positive is it will allow families to spend more time together. We shouldn’t underestimate this benefits. Having a partner who travels a lot for work, especially when you’re raising a family, can’t be a picnic. It’s likely that it puts a strain on a fair proportion of relationships.

The final major positive has to be around the effect it will have on the climate. The less we travel, in theory at least, the less we will pollute. That’s not just a benefit in the sense of climate change either. Irrespective of how you may feel about extreme weather events, the respiratory effects of pollution at a local level are difficult to challenge.

Are there any down sides though? I think the key one is that we lose the benefit of “getting out of the office”. They say a change is as good as a rest, and this can sometimes be the case when you travel for work. This is especially true if that travel is aimed at personal development, as we can come back with new ideas to improve our workplaces.

There is also the question of loneliness. We socialise online more and more. If workplaces begin to move online at a faster pace, we risk having a populace that rarely ever interacts with another human being face-to-face. We probably don’t yet know what psychological effects that may have on individuals.

Should we take up the benefits of moving more of the work we do online, we must be careful not to lose the benefits of some of things we’ve traditionally done offline. That said, Covid-19 gives us one reason to really consider what is best done online. I’d be very surprised if businesses, at the least, aren’t considering the advantages of moving online.

Free will, sense of self and AI

A couple of ideas came to me this morning around the development of human-like AI.

The first thing that came to mind is around neuroscience. Has there been a study into exactly how many states an individual neuron can achieve? I recall from my first degree that neurotransmitters are released into synaptic clefts, and that action potentials deal with transmission within a neuron. This leads me to the questions:

1. What are the maximum number of connections between a neuron and other neurons?
2. Can an action potential affect the behaviour of a neuron based on its strength (or does it work more like a PC –  with on/off or 0/1 values)?
3. What are the maximum number of neurotransmitters that a neuron can use and receive?

The reason for these questions is that they could potentially be pertinent to the development of human-like AI via neural networks. They might also be a measure of the success of an AI developed using techniques like machine learning or genetic programming (though I need to understand more about all of these to know for sure).

The second idea that occurred to me this morning was around how we conceptualise a human-like AI. Plenty of philosophers and neuroscientists seem to be coming to the conclusion that free will is essentially nothing more than a convenient comfort blanket that helps societies function. What if the same is true of our sense of self?

Perhaps our understanding that we exist is nothing more than a convenient illusion? Perhaps coming to the conclusion that we exist was nothing more than evolutionary happenstance that conferred an advantage at a specific time? Perhaps sense of self is more of a tradition that we pass on through the generations by rote, rather than something that exists as a result of a human capacity for higher level reasoning? If so, might the final stage of creating a human-like AI be simply telling it that it exists, or coding in this belief (like Asimov’s three laws)?

If so, we might actually be much closer to the existence of a human-like AI than we have heretofore realised. This line of thought led me to wondering if, with the exception of a concept of its own existence, we might consider the internet itself to be our first successful AI? After all, it takes external inputs, forwards them to the correct part of its body, generates a response, then sends that response to the appropriate part of its body. At a very basic level, that sounds quite human.

Relative pain response in Autism and Fibromyalgia

One of the things I’ve noted that Autism and Fibromyalgia seem to have in common is increased pain response. I have to admit upfront that my knowledge of Autism and Fybromyalgia is either outdated, or based on observation and use of Google. That said, I find both fascinating, and wondered if the pain aspects are being studied.

With that in mind, I produced a graph of a hypothetical experience of pain for a neurotypical person, an autistic person and a person suffering Fibromyalgia. The next stage is to find out how closely this graph fits with up-to-date research. One bonus of being a student is access to journals, which I intend to take full advantage of.

Assuming I find the time to do this properly, I may even consider writing a longer piece on it. I shall also consider how my degree might allow me to contribute to assisting with research on both Fibromyalgia and Autism, especially producing software to support said research.

 

Developing Fake News and automated research

I had another idea for a potential programming project that also led to me developing the fake news detection idea a bit. First the idea.

Currently, researchers formulate hypotheses that they then test by running experiments or analysing data. I suspect that research that involves big data has the potential to cut humans out of the loop completely. To do this, one would need to build a system capable of producing hypotheses, testing them, analysing the results for statistical significance, and thus formulating new hypotheses to test. This feels like something that machine learning might be perfectly suited to.

From an ethical point of view, it would probably need to be limited to hypotheses that can be tested using pre-existing data. I can imagine a scenario, for example, where a system might spread word of a fake virus to test how people react. This would be highly unethical, and potential harmful. This is probably another idea that is a little big for a degree project too.

On my fake news detection idea, perhaps a better application would be attempting to delineate facts from opinion. Most respectable news sources attempt to do this, but not all do. A system would need to be trained to recognise patterns of words that correspond to the presentation of facts and of opinion. This might include analysing links within content, perhaps with the system containing a list of respected journals and research organisations. That said, linking to these would be no guarantee that a piece wasn’t based on the opinion of the writer on the originating research.

 

A couple more ideas for programming projects

Two more ideas that have come to mind are:

A memetic prediction algorithm. The aim of this would be to use machine learning to predict the spread of an idea through a population. For this to work, it would need to be able to take an idea presented in image/text/video/audio form and identify the probability that it will achieve success. Ideally it might also suggest improvements, though this would be a secondary aim. The key considerations would be defining success, identifying factors that lead to success and identifying a dataset that can be used to allow a machine learning system to build an appropriate predictive model.

An evolutionary programme replication system. My observations of AI and robotics thus far have been that the focus has been predominantly on outcome. Systems and products are built and designed with a specific purpose in mind, whereas evolution has tended to rely on a combination of happenstance mutation and successful replication of individual genes. Whilst I’m sure there are people working on replicating genetic processes within programming, I would be keen to attempt this too. The aim would be to create a group of simple programmes with the sole aim of replication and the potential to combine in unique ways. A sort of genetic API. The key issue here is my own lack of subject knowledge. It has been almost 20 years since I studied genetics formally. I also suspect such a project would be too large for a degree level course. It would however be incredibly fascinating to attempt.

A couple of ideas for programming projects

As part of my discussions with a friend about completing my Computing and IT degree, he suggested that now was a great time to be thinking about what projects I might be interested in working on when I complete the computing and IT project in a few years time. I totally agreed and, aware of my tendency to write down ideas on scraps of paper that I later lose, decided to use the internet to prevent their loss.

So, with that in mind, my first couple of ideas are:

A fake news browser tool that rates news content on likelihood to be unbiased. This is really expanding on fact checking and newspaper bias websites. The initial flaw I see with it is that those that need it are probably least likely to want it. That said, it could potentially be a tool to tackle the increasingly common usage of outright lies in the media by politicians, which could be seen as damaging our democracy.

A criminal patterns in search tool. The idea here is to create a tool that can use search data to spot patterns of behaviour that may signify an individual is planning to commit a crime. The immediate issue here is how to preserve anonymity, whilst also being able to identify a potential criminal. I could see this being a project that would benefit from the expertise of David Canter, the eminent Forensic Psychologist who wrote Mapping Murder. I could also see this being regarded as “pre-crime” as per Minority Report, with the same potential pitfalls.

Hopefully I’ll come up with better ideas as I learn more and do a bit more research.

 

Thoughts on Robbie

As part of my second degree (Computing and IT with The Open University… I graduate in 2024), I have to complete various written activities outside of the main assignments. I’ve tended to quickly rush through these (bar the Maths ones, which are HARD), but I recently had a chat with an old friend that made me reconsider this. With that in mind, I’m going to try to complete these tasks online (barring any that are assessed as part of the degree… don’t want to encourage plagiarism).

So, my first attempt at this is to answer questions about the short story Robbie, in Isaac Asimov’s I Robot (which I really wish I’d read before now).

a. Who are the main characters in this story? Give a brief summary of the plot.

Gloria, her parents and Robbie, a robot playmate. Essentially, Gloria’s Mum is a technophobe and conspires to get rid of her robot. Gloria is distraught to the point of depression. Eventually her father conspires to return said robot, in the process of which the robot saves Gloria’s life. At this point, Gloria’s Mum grudgingly accepts Robbie back into the home.

b. What are the main points being made?

For me the main point is about how robots can be both an object of fear and of love. The Mum fears what she cannot understand, while Gloria anthropomorphises Robbie to the point that she ignores other children. Perception is key here, and will need to be considered as more complex robots enter our everyday lives.

c. To what extent do you think a playmate robot such as Robbie is likely to exist in the future?

Such robots already exist to a degree, though they are not as complicated. They are already being used to support autistic children, though I believe there’s not yet a great deal of evidence as to their efficacy. As the price of building them comes down, they are likely to become ever more complex and popular.