## A NEW ENCOUNTER WITH THE HUMAN GENOME

The human genome was announced "completely solved" in 2003. This announcement came exactly 50 years after the publication of the seminal paper on the structure of DNA. The data collected in the human genome sequencing project made it possible to start a new area of research called "Genomics". It took only several years to realize that our earlier view of our genome was rather na?ve. In fact, many "text book facts" had to be revised. In this talk I will explain some of these unexpected recent discoveries.

## A NEUROECONOMICS APPROACH TO THE MATCHING LAW

According to Herrnstein's matching law, the frequency of choosing an alternative in a repeated-choice experiment is equal to the fraction of rewards obtained from that alternative. I will discuss three questions concerning the matching law, which range from synaptic physiology to economic theory. (1) What is the neural basis of the matching law? I will show that matching behavior naturally emerges if changes in synaptic efficacies in the brain are proportional to the covariance of reward and neural activity. (2) Is there a normative theory of matching?

## Internal no-regret with imperfect monitoring (joint with Ehud Lehrer)

Abstract:We define the notion of internal regret-free strategies in sequential decision problems with imperfect monitoring, and prove that an internal regret-free strategy exists.We show that if all the players in a repeated game with imperfect monitoring play an internal regret-free strategy, each one takes the other players as nature, then the empirical distribution on the entries of the matrix coonverges to the set of partially specified correlated equilibria.

## ערב ספרים

**הדוברים:**עדנה אולמן- מרגליתאוה אילוזאוריאל פרוקצ'יהאריאל רובינשטייןהאירוע פתוח לכל - אפשר להביא חברים.

## THE DEVELOPMENTAL CONSTRUCTION OF HEREDITY

The inheritance of developmentally triggered variations - phenotypic variations that are independent of variations in DNA sequence, and DNA changes that are guided by epigenetic systems ? is now recognized to contribute to the transmission of information between generations of organisms, as well as being a crucial part of their ontogeny. However, such hereditary transmission has not yet been fully incorporated into evolutionary theory.

## GETTING MORE OUT OF MATHEMATICS THAN WE PUT IN

Mathematical concepts have often proved to have more in them than in their original definition. Some of this "latent information," originally ignored by mathematicians and physicists alike, has proved to be the key to very deep and surprising properties of the physical universe. The views of Penrose, Plato, and Descartes will be discussed. Penrose's 1098 page book, The Road to Reality, will be used as a prop in a topological experiment.

## AN AXIOMATIC CHARACTERIZATION OF THE THEIL INEQUALITY ORDERING

We provide a characterization of the Theil ordering of income inequality that uses ordinal axioms only.

## GETTING MORE OUT OF MATHEMATICS THAN WE PUT IN

Mathematical concepts have often proved to have more in them than in their original definition. Some of this "latent information," originally ignored by mathematicians and physicists alike, has proved to be the key to very deep and surprising properties of the physical universe. The views of Penrose, Plato, and Descartes will be discussed. Penrose's 1098 page book, The Road to Reality, will be used as a prop in a topological experiment.

## A STUDY IN THE PRAGMATICS OF PERSUASION: A GAME THEORETICAL APPROACH (JOINT WORK WITH KOBI GLAZER)

A speaker wishes to persuade a listener to take a certain action.

## BACKWARD INDUCTION AND COMMON STRONG BELIEF OF RATIONALITY

In 1995, Aumman showed that with appropriate definitions, common knowledge of rationality (CKR) in games of perfect information (PI) implies that the backward induction (BI) outcome is reached. His work was criticized because it assumes rationality also at unreached vertices of the game tree. Here we avoid the criticism by replacing "common knowledge" by "strong belief" where a player strongly believes a proposition p if he believes p at every v that is logically consistent with p. The relation to previous work of Battigalli and Siniscalchi is discussed.