Artificial Intelligence and Functional Programming
9 views | +0 today
Follow
Artificial Intelligence and Functional Programming
Artificial intelligence has been a large area of study since its inception in the 1950's.
What is intelligence? How can it be quantified, studied, and implemented on machines? How can machines model human, or higher, intelligence? These are the questions that AI researchers have been trying to answer.
Functional programming is a programming paradigm based predominantly on the lambda calculus of Alonzo Church, and centered around beliefs that mathematical purity will allow for better compiler optimisation and reasoning about correctness. This declarative approach to programming has been picking up over the last few years, and I consider functional programming one of my favourite activities. I am personally fond of the Haskell language. Functional programs tend to be shorter, and with fewer bugs. Functional programming is particularly applicable to parallel computing because it eschews unsafe side effects, which cause the majority of difficulty in parallelism.
Curated by George Wilson
Your new post is loading...
Your new post is loading...
Scooped by George Wilson
Scoop.it!

Tackling the Awkward Squad

George Wilson's insight:

this paper covers a monadic system for modelling IO in a purely functional environment. It is of great significance to the Haskell community, but I believe that as our comptuers get more and more cores, and side effects choke our ability to write parallel algorithms, even abstract concepts like monads will find a place among mainstream programming, as a safe way of modelling solutions to our problems.

more...
No comment yet.
Scooped by George Wilson
Scoop.it!

Artificial intelligence: Job killer or your next boss? | ZDNet

Artificial intelligence: Job killer or your next boss? | ZDNet | Artificial Intelligence and Functional Programming | Scoop.it
As software automates an increasing number of tasks, is it time to reassess traditional workplace roles and create an office that suits both man and machine?
George Wilson's insight:

This article discusses the interesting topic of AI being integrated into previously entirely human - in this case business - systems. We will be seeing more and more of this as AI continues to mature, and more becomes possible.

more...
No comment yet.
Scooped by George Wilson
Scoop.it!

High Scalability - High Scalability - Prismatic Architecture - Using Machine Learning on Social Networks to Figure Out What You Should Read on the Web 

High Scalability - High Scalability - Prismatic Architecture - Using Machine Learning on Social Networks to Figure Out What You Should Read on the Web  | Artificial Intelligence and Functional Programming | Scoop.it
This post on Prismatic ’s Architecture is adapted from an email conversation with Pri...
George Wilson's insight:

Computation is being used to enrich every aspect of our social lives. Websites such as Facebook target ads to their users based on their 'likes' and keywords from their status updates and comments. Using machine learning algorithms, this takes that a step further.

more...
No comment yet.
Scooped by George Wilson
Scoop.it!

Introduction to Concurrent Haskell

Introduction to Concurrent Haskell | Artificial Intelligence and Functional Programming | Scoop.it
Simon Marlow introduces some of the main features of Concurrent Haskell: forking threads, MVars, asynchronous I/O, simple inter-thread protocols.
George Wilson's insight:

Purely functional programming, as implemented in the Haskell programming language, allows for elegant and dependable concurrent and parallel code, with excellent tools for reasoning about program correctness. I believe the ideas cultivated in the Haskell community will penetrate mainstream langauges to better enable them to reason in an inherently parallel environment.

more...
No comment yet.
Scooped by George Wilson
Scoop.it!

Watson goes to college: How the world's smartest PC will revolutionize AI

Watson goes to college: How the world's smartest PC will revolutionize AI | Artificial Intelligence and Functional Programming | Scoop.it
IBM’s Watson made waves in AI when it defeated a pair of champions on Jeopardy! Now, engineering students at RPI, Watson’s new home, hope to use the supercomputer to gain fresh insights into machine learning.
George Wilson's insight:

Natural language processing is a particularly computationally intensive task (we humans are only so adept because the human brain is optimised for this task).

Watson represents a monumental achievement in the field of AI nautral langauge processing. It defeated the two greatest human Jeaopardy! players at the popular game.

more...
No comment yet.
Scooped by George Wilson
Scoop.it!

Have Google's Self-Driving Cars Hit A Technological Roadblock? - Forbes

Have Google's Self-Driving Cars Hit A Technological Roadblock? - Forbes | Artificial Intelligence and Functional Programming | Scoop.it
Google driverless car operating on a testing path (Photo credit: Wikipedia) Henry Blodget has an interesting piece at Business Insider about challenges Google is facing in it's quest to create a self-driving car based on conversations with...
George Wilson's insight:

Self-driving cars offer an interesting challenge to AI researchers -- they involve many interesting topics. Machine vision, pattern analysis, the clarification of imperfect or contradictory GPS data, machine learning (adapting the ideal driving style for the car in question). I believe that this product will spur a lot of great research in the field of AI.

more...
No comment yet.