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

Data Mining Practical Machine Learning Tools and Techniques

http://www.csun.edu/~andrzej/COMP529-S05/papers/Mesh%20Networks%20Survey.pdf

Wireless mesh networks: a survey

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

We examine a potential technique of performing a classification task based on compressively sensed (CS) data, skipping a computationally expensive reconstruction step. A deep Boltzmann machine is trained on a compressive representation of MNIST handwritten digit data, using a random orthoprojector sensing matrix. The network is first pre-trained on uncompressed data in order to learn the structure of the dataset. The outer network layers are then optimized using backpropagation. We find this approach achieves a 1.21% test data error rate at a sensing rate of 0.4, compared to a 0.99% error rate for non-compressive data.

Direct classification from compressively sensed images via deep Boltzmann machine

We consider a wireless sensor network for distributed estimation over Rayleigh fading channels. The sensors transmit their observations over fading channels to a fusion center, where a source parameter is estimated. Since the sensor transmissions add incoherently over a multiple access channel, we consider partial channel knowledge at the sensors to improve performance. We calculate the variance of the estimate when the channel phase is quantized uniformly and fed back to the sensors. We show that as few as 3 bits of feedback is sufficient for a loss in performance of about 5%. We also show that the performance is robust in the presence of feedback errors.

Performance of distributed estimation over multiple access fading channels with partial feedback

This paper focuses on the interaction and integration of several critical components of a mobile ad-hoc network (MANET) using smart antenna systems. A MANET is a wireless network where the communicating nodes are mobile and the network topology is continuously changing. One of the central motivations for this work comes from the observed dependence of the overall network throughput on the design of the adaptive antenna system and its underlying signal processing algorithms. In fact, a major objective of this work is to study and document the overall efficiency of the network in terms of the antenna pattern and the length of the training sequence used by the beamforming algorithms. Furthermore, the paper also considers problems dealing with: the choice of direction of arrival algorithm; the performance of the adaptive beamformer in the presence of antenna coupling effects; combating the effects of fading channels on the overall system.

Adaptive antenna systems for mobile ad-hoc networks

Modern audio source separation techniques rely on optimizing sequence model architectures such as, 1D-CNNs, on mixture recordings to generalize well to unseen mixtures. Specifically, recent focus is on time-domain based architectures such as Wave-U-Net which exploit temporal context by extracting multi-scale features. However, the optimality of the feature extraction process in these architectures has not been well investigated. In this paper, we examine and recommend critical architectural changes that forge an optimal multi-scale feature extraction process. To this end, we replace regular $1-$D convolutions with adaptive dilated convolutions that have innate capability of capturing increased context by using large temporal receptive fields. We also investigate the impact of dense connections on the extraction process that encourage feature reuse and better gradient flow. The dense connections between the downsampling and upsampling paths of a U-Net architecture capture multi-resolution information leading to improved temporal modelling. We evaluate the proposed approaches on the MUSDB test dataset. In addition to providing an improved performance over the state-of-the-art, we also provide insights on the impact of different architectural choices on complex data-driven solutions for source separation.

Audio Source Separation via Multi-Scale Learning with Dilated Dense U-Nets

A hybrid model for speech analysis/synthesis is proposed. It relies on a time-varying autoregressive moving-average (ARMA) model and the short-time Fourier transform (STFT). The model is hybrid in that the periodic (narrowband) component in speech is represented in the frequency domain by a harmonic-based STFT, while the random component in speech is represented by a random noise sequence, appropriately shaped by the ARMA model. The time-varying ARMA model has a dual function (namely, it creates a spectral envelope that fits accurately the harmonic STFT components) and provides for the spectral shaping of random noise. This hybrid model essentially incorporates the benefits of waveform coders by employing the STFT and the benefits of traditional vocoders by using an appropriately shaped noise sequence; thus, it is expected to yield robust speech synthesis at low data rates. >

Andreas Spanias

Papers

Direct classification from compressively sensed images via deep Boltzmann machine

Performance of distributed estimation over multiple access fading channels with partial feedback

Adaptive antenna systems for mobile ad-hoc networks

Audio Source Separation via Multi-Scale Learning with Dilated Dense U-Nets

A hybrid model for speech synthesis