Education strategy

What exactly do we mean by “learning” in multi-agent context?

A natural definition is that players “learn” if they eventually succeed in predicting their opponents’ behavior with a high degree of accuracy (Foster and Young, 2001).

There is a well-known condition in statistics that guarantees that all players will learn to predict in the strong sense. Namely, it suffices that each player’s forecast of the others’ behavior, conditional on his own behavior, never exclude events that have positive probability under their actual joint behavior. This is the absolute continuity condition (Blackwell and Dubins, 1962; Kalai and Lehrer, 1993).

Interactive learning is inherently more complex than single-agent learning,because the act of learning changes the object to be learned. If agent A is trying to learn about agent B, A’s behavior will naturally depend on what she has learned so far, and also on what she hopes to learn next. But A’s behavior can be observed by B, hence B’s behavior may change as a result of A’s attempts to learn it. The same holds for B’s attempts to learn about A.

This feedback loop is a central and inescapable feature of multi-agent learning situations. It suggests that methods which work for single-agent learning problems may fail in multi-agent settings. It even suggests that learning could fail in general, that is, there may exist situations in which no rules allow players to learn one another’s behavior in a completely satisfactory sense.



THE POSSIBLE AND THE IMPOSSIBLE IN MULTI-AGENT LEARNING, H. Peyton Young, Dept. of Economics Working paper, Oxford Univ., Jan 2007








Over educated ?????

In recent years, alongside the rapid expansion in higher education, there have been important changes in the types of qualifications being awarded by universities. While many of these new qualifications have emerged in response to changing economic needs, not all of them are career related. Thus, over-educated workers could essentially be comprised of those who have non-professional qualifications, a low quality of education, or both. Pryor and Schaffer (1999) showed that US workers who experienced downward occupational mobility generally had lower cognitive skills irrespective of educational credentials.

Beyond all aspects of human capital and job characteristics, several factors may give rise to labour market rigidities that limit the capacity of the market to fully utilise and reward highly educated workers. Such constraints could arise from family commitments, regional immobility or restrictive work practices. The Newcastle Alumni Survey containing a wealth of information on individual family circumstances and personal commitments was examined to see how far they may result in some graduates taking jobs that require less than their educational credentials.

The most distinguishing feature of the Newcastle Alumni Survey is that it is the only British data set that contains two direct questions measuring the extent of education under- utilisation. The first question is: “What is/was the minimum formal qualification level required for entering this job?” and the second question is: “What do you believe to be the education level required to actually do this job?” Answers to both questions are on a four-point scale as follows: postgraduate qualification, degree, sub-degree qualification, and no qualifications required. The first question provides a match between acquired and required qualifications to get the job, whereas the second question provides a direct measure of over-education in terms of job content. As all previous studies for the UK have replied on questions framed as in the first question, the incidence of over-education may have been overestimated by past researchers for this reason.

Looking at family commitment variables, having children prior to first job decreases (at the 10 percent level of significance) the probability of being over-educated. High debt commitments (i.e., debts in excess of £1000 upon leaving the University) raise the probability of being over-educated in the first employment. Data used from the Newcastle Alumni Survey was especially commissioned for research into over-education in the graduate population, to employ a subjective measure based upon a question that measures the education level needed to do the job. Our descriptive statistics revealed that about one in five university graduates were not employed in graduate- level positions after spending some time in the labour force. Unobserved ability differences between similarly educated workers and the quality of education are usual conjectures offered to explain these findings.

Source:The Determinants of Graduate Over-Education,
Peter Doltona and Mary Sillesb
a)Department of Economics, University of Newcastle, Newcastle upon Tyne, NE1 7RU, England.
b)St. John’s College, University of Oxford, Oxford, OX1 3JP, England.








In the field of education, the vitality of institutions and organisations are function of vitality of teachers, administrators and other stakeholders, hence, the vitality of the students is directly related to the vitality of schools and institutions of higher learning. Teachers need to recognise the responsibility for helping to prepare every student for followings:
- look forward to a future that will bring many changes and to accept the responsibility for helping to shape that future.
- recognise his special aptitudes and abilities, and thus facilitate the development of his individuality,
- develop a sound basis for accepting a defensible and evolving system of moral and ethical values for guidance in exercising his responsibilities as a citizen,
- recognise that happiness comes primarily from progress in achieving and helping others to achieve worthwhile goals and objectives,
- seek and utilize learning and knowledge as a basis for understanding meanings in relationship to his life and to society,
- learn to use the scientific method as a basis for studying and resolving the problems he encounters.