Learning from examples with unspecified attribute values

A challenging problem within machine learning is how to make good inferences from data sets in which pieces of information are missing. While it is valuable to have algorithms that perform well for specific domains, to gain a fundamental understanding of the problem, one needs a “theory” about how to learn with incomplete data. The important contribution of such a theory is not so much the specific algorithmic results, but rather that it provides good ways of thinking about the problem formally. In this paper we introduce the unspecified attribute value (UAV) learning model as a first step towards a theoretical framework for studying the problem of learning from incomplete data in the exact learning framework.In the UAV learning model, an example x is classified positive (resp., negative) if all possible assignments for the unspecified attributes result in a positive (resp., negative) classification. Otherwise the classification given to x is “?” (for unknown). Given an example x in which some attributes are unspecified, the oracle UAV-MQ responds with the classification of x. Given a hypothesis h, the oracle UAV-EQ returns an example x (that could have unspecified attributes) for which h(x) is incorrect.We show that any class of functions learnable in Angluin’s exact model using the MQ and EQ oracles is also learnable in the UAV model using the MQ and UAV-EQ oracles as long as the counterexamples provided by the UAV-EQ oracle have a logarithmic number of unspecified attributes. We also show that any class learnable in the exact model using the MQ and EQ oracles is also learnable in the UAV model using the UAV-MQ and UAV-EQ oracles as well as an oracle to evaluate a given boolean formula on an example with unspecified attributes. (For some hypothesis classes such as decision trees and unate formulas the evaluation can be done in polynomial time without an oracle.) We also study the learnability of a universal class of decision trees under the UAV model and of DNF formulas under a representation-dependent variation of the UAV model.     

Accessing the spoken word

Spoken-word audio collections cover many domains, including radio and television broadcasts, oral narratives, governmental proceedings, lectures, and telephone conversations. The collection, access, and preservation of such data is stimulated by political, economic, cultural, and educational needs. This paper outlines the major issues in the field, reviews the current state of technology, examines the rapidly changing policy issues relating to privacy and copyright, and presents issues relating to the collection and preservation of spoken audio content .     

The Value of Temptation

There is an implicit assumption in electronic commerce that induces the buyers to believe that their deals will be handled appropriately. However, after a seller has already committed to a buyer, he may be tempted by several requests though he will not be able to supply them all. We analyze markets in which a finite set of automated buyers interacts repeatedly with a finite set of automated sellers. These sellers can satisfy one buyer at a time, and they can be tempted to break a commitment they already have. We have found the perfect equilibria that exist in markets with a finite horizon, and with an unrestricted horizon. A significant result stemming from our study reveals that sellers are almost always tempted to breach their commitments. However, we also show that if markets' designers implement an external mechanism that restricts the automated buyers actions, then sellers will keep their commitments.     

Segmentation of Colour Images with Highlights and Shadows sing Fuzzy-like Reasoning

In this paper, a fuzzy-like technique is presented that resolves several difficult issues related to image segmentation, such as highlights and shadows. Large, relatively continuous, areas within an image are usually easy to segment, and the pixels included within different segments are often determined by using derived edge information. However, in many cases, pixels which lie between segments or in high frequency areas of an image cannot be easily categorised as belonging to any particular segments. Typically, according to the dichromatic reflection model, these pixels may belong to the matte, highlight or shadow area of the closest segment; or, in association with neighbouring pixels, they make up a separate smaller segment. The dichromatic reflection model is applied here to merge highlight and shadow areas with matte areas in an image. By segmenting those pixels into proper regions, the proposed fuzzy-like reasoning approach provides a more human-like segmentation of images.     

Sialic acid in lymphocytes and sera from mothers and their newborns after normal deliveries

Sialic acid was determined in newborn and maternal lymphocytes as well as in newborn and maternal serum immediately after normal deliveries. A significant increase in sialic acid in newborn as compared to maternal lymphocytes was found. Sialic acid concentration seems to be sex-dependent for newborn lymphocytes only, and higher concentrations of sialic acid were observed in male as compared to female newborn lymphocytes. No differences between the sialic acid concentration in maternal and control lymphocytes could be found. In newborn serum, sialic acid was significantly reduced as compared to maternal or control serum. No sex-linked differences were observed. No differences between the concentration of maternal and control serum were detected.     

S-adenosylmethionine metabolism and DNA methylation in hydrazine-treated rats

The treatment of rats with hepatotoxic doses of hydrazine (NH2-NH2) induces the rapid formation of 7-methylguanine and O6-methylguanine in liver DNA. The methyl moiety in these reactions might be derived from the cellular S-adenosylmethionine pool because radioactivity administered to these rats as methionine rapidly appears in the DNA as methylated guanine. An increased incorporation of radioactivity into 5-methylcytosine was previously reported followed by subsequent suppression. This increased radiolabeling of 5-methylcytosine coincided with time of maximal DNA guanine methylation. To determine the nature of S-adenosylmethionine metabolism during the period of DNA methylation induced by hydrazine treatment, and to determine if the increased radiolabeling of 5-methylcytosine at this time reflected an actual increase in 5-methylcytosine synthesis, liver DNA synthesis and S-adenosylmethionine levels and turnover were assayed. Liver S-adenosylmethionine concentrations varied slightly between control rats and hydrazinetreated rats during the first five hours after hydrazine administration, and no difference was detectable in the incorporation of administered [3H]methionine into S-adenosylmethionine. Because S-adenosylmethionine specific radioactivity in hydrazine-treated rats was not different from control rats, the previously observed increased radiolabeling of 5-methylcytosine appeared to represent an actual increase in synthesis. This conclusion was supported by finding that incorporation of radioactive thymidine into DNA was also accelerated immediately following hydrazine administration, again followed by a decrease. 5-Methylcytosine sythesis, therefore, appears to follow DNA synthesis during hydrazine toxicity, and formation of 7-methylguanine and O6-methylguanine in liver DNA of hydrazine-treated rats occurs during a short period of increased DNA sythesis and 5-methylcytosine formation very early in hydrazine toxicity.     

Sensation seeking, augmenting-reducing, and absolute auditory threshold: A strength-of-the-nervous-system perspective.

Tested 40 male and female high school and college students on the General and the Disinhibition subscales of the Sensation Seeking Scale (Form IV), the Reducer-Augmenter Scale (RAS), and the absolute auditory threshold (AAT). General sensation seeking correlated significantly with the RAS and the AAT. Both results were general across sex. These findings, that high-sensation seekers tend to be reducers and to lack sensitivity to weak stimulation, are interpreted as supporting strength-of-the-nervous-system theory more than the formulation of M. Zuckerman et al (see record 1975-02479-001). (36 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Issues and Techniques in Networked-Based Distributed Healthcare: Overview

For telemedicine to reach its potential in the Internet2 environment,significant progress must be made in a number of areas. The emergingtelecommunications will include the Internet2 between medicalinstitutions and emerging broadband loop and wireless technologiesfor interconnections to homes. Within the technical environment,problems that must be addressed include the development of robotic-controlled,remote-monitoring devices unique to the various medical specialties;the encoding, transmission, and decoding of raw data specificto the specialized needs of various medical application areas;and the reduction of data for storage and retrieval in patientrecords. Institutionally, human factor and information accessissues with respect to physicians, patients, and non-physicianhealthcare providers must be addressed; healthcare insuranceprovider policies, government regulation, and state licensinglaws must be adapted for a new paradigm of healthcare delivery;and the philosophy of capital investment for the delivery ofhealthcare must be revised.Many current practiceswill be supplanted by home healthcare alternatives, the use ofpatient-initiated preventive healthcare initiatives, and remote-clinicactivity. These applications will result in central record archivingsystems; the need for encryption schemes, virus protection, andvandalism protection; extensive, additional connectivity betweenhealthcare providers and professionals; and new commercial playersin the healthcare arena.The hot applicationareas will include women's health, children's health, healthcarefor senior citizens, home healthcare, well-patient monitoring,and early-detection of medical problems.     

Practical voting rules with partial information

Voting is an essential mechanism that allows multiple agents to reach a joint decision. The joint decision, representing a function over the preferences of all agents, is the winner among all possible (candidate) decisions. To compute the winning candidate, previous work has typically assumed that voters send their complete set of preferences for computation, and in fact this has been shown to be required in the worst case. However, in practice, it may be infeasible for all agents to send a complete set of preferences due to communication limitations and willingness to keep as much information private as possible. The goal of this paper is to empirically evaluate algorithms to reduce communication on various sets of experiments. Accordingly, we propose an iterative algorithm that allows the agents to send only part of their preferences, incrementally. Experiments with simulated and real-world data show that this algorithm results in an average of 35% savings in communications, while guaranteeing that the actual winning candidate is revealed. A second algorithm applies a greedy heuristic to save up to 90% of communications. While this heuristic algorithm cannot guarantee that a true winning candidate is found, we show that in practice, close approximations are obtained.     

Properties of chemically vapor-deposited Tungsten thin films on silicon wafers

Tungsten thin films were deposited onto silicon wafers by the thermal decomposition of W(CO)₆. The non-corrosive nature of W(CO)₆ and its decomposition products, tungsten and CO, allows deposition without damage to the wafer surface. The deposition rate is dependent on such parameters as the wafer temperature and the pressure of W(CO)₆. The effects of annealing on film characteristics were studied.The tungsten films show good uniformity and adhesion after deposition and also after annealing. The resistivity depends on the film thickness and, for a film 2200 Å thick, is 15 x 10^-5 ω cm. Wafer annealing is done at 800–900°C under vacuum in an atmosphere of forming gas. After the films have been annealed for 30 min, the resistivity decreases to 1.5 x 10^-5 ω cm.Auger analysis of the films shows that some CO is trapped in the tungsten matrix after deposition at a ratio of about one CO molecule to every four tungsten atoms. Annealing for 30 min drives off the CO, reducing the ratio to less than one CO molecule per 30 tungsten atoms.