Ai

Regularization is a fundamental concept in Machine Learning (#ML) and is generally used with activation functions . It is the key technique that help with overfitting. Overfitting is when an algorithm or model ‘fits’ the training data too well - it seems to good to be true. Essentially overfitting is when a model being trained, learns the noise in the data instead of ignoring it. If we allow overfitting, then the network only uses (or is more heavily influenced) by a subset of the input (the larger peaks), and doesn’t factor in all the input. ...

Neural Networks, today, help in a great set of tasks, that until very recently wasn’t possible at all - be it from computer vision, to medical diagnosis, to speech translation and forms a key cornerstone to a lot of ‘magic’ that Machine Learning and AI offers today. I did blog about Neural Network types (and MarI/O) sometime back ; I surely cannot take credit for creating these three cheat sheets but they are awesome and hope you get to use and enjoy them too. ...

Thought of the week: Artificial Intelligence stands no chance against natural Stupidity. #ArtificalIntelligence

Seem like my training data for the car - perhaps a hint of #bias. 😂 #GeekyJokes #ML #AIJokes

Neural networks have a very interesting aspect – they can be viewed as a simple mathematical model that defines a function. For a given function $f(x)$ which can take any input value of $x$, there will be some kind a neural network satisfying that function. This hypothesis was proven almost 20 years ago (“ Approximation by Superpositions of a Sigmoidal Function ” and “ Multilayer feedforward networks are universal approximators ”) and forms the basis of much of #AI and #ML use cases possible . ...

I was looking at something else and happen to stumble across something called Netron , which is a model visualizer for #ML and #DeepLearning models. It is certainly much nicer than for anything else I have seen. The main thing that stood out for me, was that it supports ONNX , and a whole bunch of other formats (Keras, CoreML), TensorFlow (including Lite and JS), Caffe, Caffe2, and MXNet. How awesome is that? ...

Trained a model to create a synthetic sound that sounds like me. This is after training it with about 30 sentences - which isn’t a lot. To create a synthetic voice, you enter some text, using which is then “transcribed” using #AI and your synthetic voice is generated. In my case, at first, I had said AI, which was generated also as “aeey” (you can listen here ). So for the next one, changed the AI to Artificial Intelligence. ...

Over the last few weeks, I built a self-driving car - which essentially is a remote control Rx car that uses a raspberry pi running Python, TensorFlow implementing a end-to-end convolution neural network (CNN) Of course other than being a bit geeky, I do think this is very cool to help understand and get into some of the basic constructs and mechanics around a number of things - web page design, hardware (maker things), and Artificial Intelligence principles. ...

Can #AI make me look (more) presentable? The jury is out I think. This is called style transfer, where the style/technique from a kind of painting (could be a photos too) is applied to an image, to create a new image. I took this using the built-in camera on my machine sitting at my desk and then applying the different kind of ‘styles’ on it. Each of these styles are is a separate #deeplearning model that has learned how to apply the relevant style to a source image. ...

Yesterday it worked Today it is not working #AI is like that #Haiku #GeekyHaiku #GeekyJokes

One of the challenges we have with AI is that there isn’t any universal definition - it is a broad category that means everything to everyone. Debating the rights, and, the wrongs, and the should’s and the shouldn’t s is another post though. DARPA outlines this as the “programmed ability to process information” and across a certain set of criteria that span across perceiving, learning, abstracting, and, reasoning. ...

Cognitive Bias

Thinking about #machinelearning? It will be helpful to understand some numerical computations and concepts that affect the #ML algorithm. One might not interact with these directly, but we surely can feel the effect. The things you need to think about are: 1. Overflow and underflow - thinking of them as rounding up or down errors that shift the functions enough, and compounded across the iterations cam be devastating. Of course can also easily get to division by zero. ...

I know I have had to explain this a lot in most #AI related conversations that I have had - and lately those have been quite a lot. In my experience, most people use these terms interchangeably when they are meaning one over the other. Whilst they all are (inter)related and one might help trigger the other, they are still fundamentally different and at some point, it is good to understand the differences. I like the image below ( source ) that whilst on one hand is showing a time graph, the correlation between them and how one is a subset of the other is what is interesting. ...

There of course are many, but for someone coming from computer science, and, software engineering, where the environment is relatively clean and certain (deterministic), it usually is a leap to understand that Machine Learning (and other elements of #AI) are not. Machine learning, is based on probability theory and deals with stochastic (non-deterministic) elements all the time. Nearly all activities in machine learning, require the ability to factor and more importantly, represent and reason with uncertainty. ...

When it comes to AI and the limits of human kind, what better example that shows the art of the possible than what Microsoft is doing with special awareness and HoloLens and other sensors. And not only can this replay time and allow you to have a ’living memory’ but it also is mobile. ...

Of course you heard of Neural Networks! In the context of #AI they are all the buzz of course. You might have heard of some such as DFF (Deep Feed Forward) or RNN (Recurrent neural networks)? Or perhaps you meant Recursive neural networks? Irrespective, it can be quite messy as you can see below and it would be somewhat important to have some understanding of the differences. And in case you are thinking, well what good or use is all this? Here is one example ( MarI/O - Machine Learning for Video Games) that shows how a computer learned to play Mario using DeepMind and a Neural network. ...

Continuing the previous #ArtificialIntelligence theme. Wanted to see what and how does Amazon’s rekognition work and different from the #AI offerings from the others, such as Microsoft. Here is a #ProjectMurphy image’s confidence score. I am glad to see that there is a 99% confidence that this is a person. Object and Scene detection The request POST is quite simple: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 { "method": "POST", "path": "/", "region": "us-west-2", "headers": { "Content-Type": "application/x-amz-json-1.1", "X-Amz-Date": "Thu, 01 Dec 2016 22:21:01 GMT", "X-Amz-Target": "com.amazonaws.rekognitionservice.RekognitionService.DetectLabels" }, "contentString": { "Attributes": [ "ALL" ], "Image": { "Bytes": "..." } } } And so is the response: ...

So, been spending a lot of time recently around many things related to Artificial Intelligence (#AI). More on that some day. :) Was curious about yesterdays Amazon’s announcement to jump on this bandwagon. Of course Microsoft and others have been there. I don’t know to what extend has Amazon been working on this, but given Alexa has been out for a couple of years, I know they have had rich pickings of tuning this further. ...