Herself’s Artificial Intelligence

If you are interested in network information theory you might want to check out this pdf of combined lecture notes from several graduate classes.

Network information theory deals with the fundamental limits on information flow in networks and optimal coding techniques and protocols that achieve these limits. It extends Shannon’s point-to-point information theory and the Ford–Fulkerson max-flow min-cut theorem to networks with multiple sources and destinations, broadcasting, interference, relaying, distributed compression and computing. Although a complete theory is yet to be developed, several beautiful results and techniques have been developed over the past forty years with potential applications in wireless communication, the Internet, and other networked systems.
This set of lecture notes, which is a much expanded version of lecture notes used in graduate courses over the past eight years at Stanford, UCSD, CUHK, UC Berkeley, and EPFL, aims to provide a broad coverage of key results, techniques, and open problems in network information theory. The lectures are organized in a “top-down” manner into four parts: background, single-hop networks, multi-hop networks, and extensions. The organization attempts to balance the introduction of new techniques and new models. Extensions (if any) to many users and large networks are discussed throughout. The lectures notes provide a unified, simplified, and formalized treatment of achievability using a few basic lemmas. The proofs in the lecture notes use elementary tools and techniques, which should make them accessible to graduate students in EE, CS, Statistics, and related fields as well as to researchers and practitioners in industry.

Network Information Theory Notes

pdf only

JBotSim is a simulation library for prototyping distributed algorithms in dynamic networks. The style of programming in JBotSim is event-driven: your algorithms are defined as subroutines to be executed when some particular events occur (e.g. ring of an alarm clock, appearance/disappearance of a link, arrival of a message, movement of the underlying node, etc.). Movements of the nodes can be either controlled by the algorithm itself (e.g. mobile robots), by an independent algorithm (e.g. mobility model), or by means of mouse-based interactions during the execution. Besides its features, the main asset of JBotSim is its simplicity of use. ( read more The JBotSim Library )

More information:
JBotSim Library at Source Forge
Examples

Pyevolve was developed to be a complete genetic algorithm framework written in pure python, the main objectives of Pyevolve is:

* written in pure python, to maximize the cross-platform issue;
* easy to use API, the API must be easy for end-user;
* see the evolution, the user can and must see and interact with the evolution statistics, graphs and etc;
* extensible, the API must be extensible, the user can create new representations, genetic operators like crossover, mutation and etc;
* fast, the design must be optimized for performance;
* common features, the framework must implement the most common features: selectors like roulette wheel, tournament, ranking, uniform. Scaling schemes like linear scaling, etc;
* default parameters, we must have default operators, settings, etc in all options;
* open-source, the source is for everyone, not for only one.

More information:
Blog for Pyevolve
Documentation and downloads
GA Based Sorting using Pyevolve

The sight of a cockroach scurrying for cover may be nauseating, but the insect is also a biological and engineering marvel, and is providing researchers at Oregon State University with what they call “bioinspiration” in a quest to build the world’s first legged robot that is capable of running effortlessly over rough terrain.

If the engineers succeed, they may owe their success to what’s being learned from these insects and other animals, such as the guinea hen, that have their own remarkable abilities.

. . .

Within certain limitations, Schmitt said, cockroaches don’t even have to think about running – they just do it, with muscle action that is instinctive and doesn’t require reflex control. That, in fact, is part of what the engineers are trying to achieve. Right now some robots have been built that can walk, but none of them can run as well as their animal counterparts. Even walking robots absorb far too much energy and computing power to be very useful.

“If we ever develop robots that can really run over rough ground, they can’t afford to use so much of their computing abilities and energy demand to accomplish it,” Schmitt said. “A cockroach doesn’t think much about running, it just runs. And it only slows down about 20 percent when going over blocks that are three times higher than its hips. That’s just remarkable, and an indication that their stability has to do with how they are built, rather than how they react.”

. . .

In a computer model, they’ve created a concept that would allow a running robot to recover from a change in ground surface almost as well as a guinea hen. They are studying how the interplay of concepts such as energy storage and expenditure, sensor and feedback requirements, and leg angles can produce recovery from such perturbations. Ultimately, a team of OSU engineers hopes to use knowledge such as this to actually build robots that can efficiently run over rough terrain without using significant computing power.

Full Press release at Oregon State

More information:
Modeling posture-dependent leg actuation in sagittal plane locomotion ( original paper )
Journal of Bioinspiration and Biomimetics ( lots of papers available at no charge )

Google’s Research Blog posted an article this week on its use of quantum computing algorithms

More information:
Training a Large Scale Classifier with the Quantum Adiabatic Algorithm
Qubuit.org, Center for Quantum Computing
Introduction to Quantum Computing
The Quantum Computer
Quantum Computing and Shor’s Algorithm
Quantum Computing Day 1, Google Tech Talk on YouTube
Quantum Computing Day 2, Google Tech Talk on YouTube
Quantum Computing Day 3, Google Tech Talk on YouTube