There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

A new aspect of cell membrane structure is presented, based on the dynamic clustering of sphingolipids and cholesterol to form rafts that move within the fluid bilayer. It is proposed that these rafts function as platforms for the attachment of proteins when membranes are moved around inside the cell and during signal transduction.

Functional rafts in cell membranes

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

In high-speed networks, such as 1-Gbps or higher networks, bandwidth measurement algorithms that utilize packet transmission/arrival intervals, such as packet trains and packet pairs, have a number of problems. First, network measurement for large bandwidth requires short packet transmission intervals, which causes a heavy load on the CPU. Second, network interface cards for high-speed networks usually employ Interrupt Coalescence (IC), which rearranges the arrival intervals of packets and causes bursty transmission of packets. In the present study, we introduce ICIM (Interrupt Coalescence-aware inline measurement), a new bandwidth measurement approach that overcomes these two problems. ICIM utilizes the data packets of an active TCP connection for the measurement. In order to determine the available bandwidth, rather than adjusting the packet transmission intervals, the TCP sender instead adjusts the number of packets involved in a burst and checks whether the inter-intervals of the bursts of corresponding ACK packets are increased or not. Simulation results show that ICIM can measure the bandwidth as high as some Gbps while requiring a number of data packets that is only 1/100 of that of the existing stream-based algorithm.

/pdf/icim-an-inline-network-measurement-mechanism-for-highspeed-4f6764wye3.pdf

ICIM: An Inline Network Measurement Mechanism for Highspeed Networks

Dynamic placement of the virtual network functions (VNFs) is one of the promising approaches to handling time-varying demands; when demands are small, the energy consumption can be reduced by placing the VNFs to a small number of physical nodes and shutting down unused nodes. If the demands becomes large, the VNFs are migrated to allocate the sufficient resources. In the dynamic placement of the VNFs, it is important to avoid a large number of migrations at each time because the migration requires a large amount of bandwidth. In this paper, we propose a new method to dynamically place the VNFs to follow the traffic variation without migrating a large number of VNFs. Our method is based on the model predictive control (MPC). By applying the MPC to the dynamic placement of the VNFs, our method starts migration in advance by considering the predicted future demands. As a result, our method allocates sufficient resources to the VNFs without migrating a large number of VNFs at the same time even when traffic variation occurs. Through simulation, we demonstrate that our method handles the time variation of the demands without requiring a large number of migration at any time slot.

Dynamic placement of virtual network functions based on model predictive control

In recent years, AQM (Active Queue Management) mechanisms, which support the end-to-end congestion control mechanism of TCP by performing congestion control at a router have been actively studied by many researchers. AQM mechanisms usually have several control parameters, and their effectiveness depend on a setting of those control parameters. Therefore, issues on parameter tuning of several AQM mechanisms have been extensively studied using simulation experiments. However in most of those studies, only a small number of simulation experiments are performed for investigating the effect of control parameters on the performance of AQM mechanisms. In this paper, we therefore statistically analyze a large number of simulation experiments using multivariate analysis, and quantitatively show how the performance of AQM mechanisms is affected by a setting of control parameters. In particular we analyze the performance of three AQM mechanisms: GRED (Gentle RED), DRED (Dynamic-RED), and SRED (Stabilized RED), all of which are variants of RED (Random Early Detection). Through several numerical examples, we clarify how control parameters of GRED, DRED, and SRED have impact on their steady state performance measures such as the average queue length and the packet loss probability. We present a few guidelines for configuring control parameters of those AQM mechanisms.

/pdf/on-control-parameters-tuning-for-active-queue-management-11vi4gfyvr.pdf

On control parameters tuning for active queue management mechanisms using multivariate analysis

We have proposed a new TCP version, called ImTCP (Inline measurement TCP), in [1]. The ImTCP sender adjusts the transmission intervals of data packets, and then utilizes the arrival intervals of ACK packets for the available bandwidth estimation. This type of active measurement in a TCP connection (inline measurement) is preferred because it delivers measurement results that are as accurate as active measurement, even though no extra probe traffic is injected into the network. In the present research, we combine a new capacity measurement function with the currently used measurement method to enable simultaneous measurement of both capacity and available bandwidth in ImTCP. The capacity measurement algorithm is essentially based on the packet pair technique, but also consider the estimated available bandwidth values for data filtering or data calculation, so that this algorithm promises better measurement results than current packet-pair-based measurement algorithms.

/pdf/a-merged-inline-measurement-method-for-capacity-and-1ydxd0uix6.pdf

A merged inline measurement method for capacity and available bandwidth

In this article, we introduce a new concept for the future information environment, called an “ambient information environment (AmIE).” We first explain it, especially emphasizing the difference from the existing ubiquitous information environment (UbIE), which is an interaction between users and environments. Then, we focus on an ambient networking environment (AmNE) which supports the AmIE as a networking infrastructure. Our approach of a biologically inspired framework is next described in order to demonstrate why such an approach is necessary in the AmIE. Finally, we show some example for building the AmNE.

Masayuki Murata

Papers

ICIM: An Inline Network Measurement Mechanism for Highspeed Networks

Dynamic placement of virtual network functions based on model predictive control

On control parameters tuning for active queue management mechanisms using multivariate analysis

A merged inline measurement method for capacity and available bandwidth

Towards Establishing Ambient Network Environment