infinity


The Truth About Bayesian Priors and Overfitting

@machinelearnbot

Have you ever thought about how strong a prior is compared to observed data? It features a cyclic process with one event represented by the variable d. There is only 1 observation of that event so it means that maximum likelihood will always assign everything to this variable that cannot be explained by other data. In the plot below you will see the truth which is y and 3 lines corresponding to 3 independent samples from the fitted resulting posterior distribution. Before you start to argue with my reasoning take a look at the plots where we plot the last prior vs the posterior and the point estimate from our generating process.


Three Original Math and Proba Challenges, with Tutorial

@machinelearnbot

While having myself a strong mathematical background, I have developed an entire data science and machine learning framework (mostly for data science automation) that is almost free of mathematics, and known as deep data science. You will see that you can learn serious statistical concepts (including limit theorems) without knowing mathematics, much less probabilities or random variables. Anyway, for algorithms processing large volume of data in nearly real-time, computational complexity is still very important: read my article about how bad so many modern algorithms are and could benefit from some lifting, with faster processing time allowing to take into account more metrics, more data, and more complicated metrics, to provide better results. It looks like f(n), as n tends to infinity, is infinitely smaller than log n, log(log n), log(log(log n))), and so on, no matter how many (finite number of) nested log's you have.


From Infinity to 8: Translating AI into real numbers

@machinelearnbot

In this episode, gnomes collect underpants and make a profit. The business plan is revealed via a slide, of course: AI offers something similar: (1) Collect data, (2) AI, (3) Profit! In an earlier article, I talked through the holy trinity of AI: the chicken (algorithms), eggs (data), and bacon (results). Think of this as a food chain: software is eating the world; software is fed by AI; and AI is fed by data.


From Infinity to 8: Translating AI into real numbers

#artificialintelligence

In this episode, gnomes collect underpants and make a profit. The business plan is revealed via a slide, of course: AI offers something similar: (1) Collect data, (2) AI, (3) Profit! In an earlier article, I talked through the holy trinity of AI: the chicken (algorithms), eggs (data), and bacon (results). Think of this as a food chain: software is eating the world; software is fed by AI; and AI is fed by data.


The Fundamental Statistics Theorem Revisited

@machinelearnbot

It turned out that putting more weight on close neighbors, and increasingly lower weight on far away neighbors (with weights slowly decaying to zero based on the distance to the neighbor in question) was the solution to the problem. For those interested in the theory, the fact that cases 1, 2 and 3 yield convergence to the Gaussian distribution is a consequence of the Central Limit Theorem under the Liapounov condition. More specifically, and because the samples produced here come from uniformly bounded distributions (we use a random number generator to simulate uniform deviates), all that is needed for convergence to the Gaussian distribution is that the sum of the squares of the weights -- and thus Stdev(S) as n tends to infinity -- must be infinite. More generally, we can work with more complex auto-regressive processes with a covariance matrix as general as possible, then compute S as a weighted sum of the X(k)'s, and find a relationship between the weights and the covariance matrix, to eventually identify conditions on the covariance matrix that guarantee convergence to the Gaussian destribution.


Three Original Math and Proba Challenges, with Tutorial

@machinelearnbot

While having myself a strong mathematical background, I have developed an entire data science and machine learning framework (mostly for data science automation) that is almost free of mathematics, and known as deep data science. You will see that you can learn serious statistical concepts (including limit theorems) without knowing mathematics, much less probabilities or random variables. Anyway, for algorithms processing large volume of data in nearly real-time, computational complexity is still very important: read my article about how bad so many modern algorithms are and could benefit from some lifting, with faster processing time allowing to take into account more metrics, more data, and more complicated metrics, to provide better results. It looks like f(n), as n tends to infinity, is infinitely smaller than log n, log(log n), log(log(log n))), and so on, no matter how many (finite number of) nested log's y


Goedel's Incompleteness Theorem and the Emergence of AI

#artificialintelligence

Some leading scientists like Sir Roger Penrose even argue that Goedel showed with his Incompleteness Theorem that today's computers can never reach human level intelligence or consciousness, that humans will always be smarter than current computers or any computer algorithm can ever be and that computers will never in the true sense of the word "understand" anything like higher level mathematics, especially not mathematics that deals with trans-finite sets and numbers. Many famous mathematicians (and physicists) created fascinating new theories and discovered deep and far reaching mathematical results. He proved this by using his famous "diagonal" construction (see pic below) that showed that any supposedly complete enumerated list of irrational or real numbers R will always miss some irrational numbers, thereby proving that a complete enumeration of the real numbers by the natural numbers is impossible. Cantor has actually shown that there are even an infinite number of ever bigger infinities by showing that the set of all subsets of any given infinite set is always substantially bigger (cannot be put into a 1-1 relation) than the set itself.


Introduction to Number Theory: Fascinating Facts and Conjectures about Primes and Other Special Numbers

@machinelearnbot

A few important unsolved mathematical conjectures are presented in a unified approach, and some new research material is also introduced, especially an attempt at generalizing and unifying concepts related to data set density and limiting distributions. Finally, we provide an algorithm that computes quantities related to densities, for a number of integer families, including prime numbers, and integers that are sum of two squares. The last section discusses potential areas for additional research, such as a probabilistic number theory, generating functions for composite numbers (possibly leading to a generating function for primes) as well as strong abnormalities in the continued fraction expansions for many constants, including for the special mathematical constants (Pi, K, e, and other transcendental numbers) mentioned in this article. Moreover, the family of limiting functions n, log n, log log n, log log log n etc.


The Fundamental Statistics Theorem Revisited

@machinelearnbot

It turned out that putting more weight on close neighbors, and increasingly lower weight on far away neighbors (with weights slowly decaying to zero based on the distance to the neighbor in question) was the solution to the problem. For those interested in the theory, the fact that cases 1, 2 and 3 yield convergence to the Gaussian distribution is a consequence of the Central Limit Theorem under the Liapounov condition. More specifically, and because the samples produced here come from uniformly bounded distributions (we use a random number generator to simulate uniform deviates), all that is needed for convergence to the Gaussian distribution is that the sum of the squares of the weights -- and thus Stdev(S) as n tends to infinity -- must be infinite. More generally, we can work with more complex auto-regressive processes with a covariance matrix as general as possible, then compute S as a weighted sum of the X(k)'s, and find a relationship between the weights and the covariance matrix, to eventually identify conditions on the covariance matrix that guarantee convergence to the Gaussian destribution.


Paysa CompanyRank: How Top Tech Companies Evolve Over Time

#artificialintelligence

A few weeks ago Lydia Dishman wrote a fantastic piece characterizing the top tech companies as defined by the "quality" of their talent using the Paysa CompanyRank algorithm and how these companies change in rank over time. Figure 1 depicts the Paysa CompanyRank time-series of Uber, Facebook, Google and Zynga over time. If a company loses those from top companies or begins hiring from less quality companies, their score (and relative ranking) will decrease. The analog of top publishers linking to other top publishers holds with talent moving from one top company to another.