Using human evaluation of the happiness of words, we analyze a diverse set of large-scale texts which reflect cultural experience including 50 years of music lyrics, millions of weblogs, and billions of status updates from Twitter. We find that happiness rises and falls with age and distance from the Earth's equator; the 2008 Presidential Election was the happiest day in the blogosphere in the last 5 years; and the written form of the English language exhibits a pro-social bias. We also investigate how happiness is related to geospatial information, demographics, and network topology. What are the happiest cities? Which words anti-correlate with obesity? Does your happiness correlate with that of your friends? This talk will discuss the findings in the context of our ongoing effort in the Computational Story Lab at UVM to develop a remote sensor of population level happiness: the 'Hedonometer'.

The Runaway Greenhouse is a climate state in which the coexistence of a condensed reservoir of a greenhouse gas (such as a liquid water ocean) with a vapor phase in the atmosphere becomes impossible, resulting in all the condensed reservoir being converted to vapor, with attendant extreme increases in planetary temperature. This state occurs when the stellar radiation absorbed by the planet exceeds a certain threshold which depends primarily on the radiative properties of the gas in question and the surface gravity of the planet. The Runaway Greenhouse is believed to account for the present hot, dry state of Venus, and also defines the inner edge of the habitable zone about stars. In this talk I will discuss a number of recent results concerning the Runaway Greenhouse, including the role of subsaturation in permitting metastable non-runaway states, the vertical structure of runaway atmospheres about red dwarf stars, and some peculiarities of moist convection in a condensible pure water vapor atmosphere. I will also discuss the question of whether the Earth itself may be in a metastable non-runaway state, which would imply that it is conceivable that very extreme increases in atmospheric carbon dioxide could trigger a runaway.

Is the university as an institution under attack? To many working within academia it appears to be so. The pages of the Chronicle of Higher Education are full of anxious and angry debate about the ever rising cost of tuition, the increase in for-profit institutions, the undermining of professionalism coupled with the growing reliance on badly paid temporary adjuncts to make up for the loss of tenure track positions, and the kind of underfunding and neglect of State universities that is best represented by the problems besetting the University of California. On the other hand, a recent article by Forbes magazine listed “University Professor” as one of the least stressful jobs, and each year a new book appears in the bestseller lists that blames the failure of US universities – and in some cases the whole economy – on lazy professors who would rather carry out selfish and frivolous research than teach students what they need to be successful entrepreneurs. While agreeing that the university is in peril, the cause and solution seem to lead in quite different directions.

How can we make sense of this debate? Particularly when, to many whose working lives are based within universities, the systematic under-funding of universities and scientific research appears both illogical and unstoppable. In this talk I describe recent research undertaken in universities in Chile and the US that sheds light on how this situation came about, and the impact it has on the autonomy of individual scientific communities, and the relationship between scientists and society.

Signal or information processing and shaping of a response (a.k.a, signal transduction, regulation, sensing) is a common function performed by organisms on all levels of organization. In this talk, we will study information processing in cellular systems to answer questions like: What are the fundamental physical limits to the fidelity of information processing set by the intrinsic fluctuations in the cellular biochemical machinery? How close the cells come to these limits? What can they do to improve the performance? After introducing the general theoretical framework, we will address these and related questions in the specific experimental context of mammalian NF-kB signaling.

Probability does not exist. At least no more so than "mass" "spin" or "charm" exist. Yet probability forecasts are common, and there are fine reasons for deprecating point forecasts, as they suggest an unscientific certainty in exactly what the future holds. What roles do our physical understanding and laws of physics play in the construction of probability forecasts to support of decision making and science-based policy? Will probability forecasting more likely accelerate or retard the advancement of our scientific understanding? Model-based probability forecasts can vary significantly with alterations in the method of data assimilation, ensemble formation, ensemble interpretation, and forecast evaluation, not to mention questions of model structure, parameter selection and the details of the available forecast-outcome archive. The role of each of these aspects is considered, in the context of interpreting the forecast as a real world probability; the cases of weather forecasting, climate forecasting, and economic forecasting are contrasted. The notion of what makes a probability forecast "good" is discussed, distinguishing "skill" and "value".

For a probability forecast to be decision-relevant (as a probability forecast), it must be reasonably interpreted as a basis for rational action through the reflection of the probability of the outcomes forecast. This rather obvious sounding requirement proves to be the source of major discomfort as the distinct roles of uncertainty (imprecision) and error (structural mathematical "misspecification") are clarified. Probabilistic forecasts can be of value to decision makers even when it is irrational to interpret them as probability forecasts. A similar statement, of course, holds for point forecasts. In this context: do decision-relevant probability forecasts exist? And if not, how might today's forecasts be augmented?

Recent experimental evidence shows that the genetic networks of living organisms operate in a critical phase, namely, at the brink of a phase transition between ordered and chaotic dynamics. This has profound implications because it is precisely at the critical phase where robustness and evolvability, two central properties of living systems, can be understood within the same conceptual framework. However, the evolutionary mechanisms by which these networks became dynamically critical are still unknown. In this talk I will present an evolutionary model showing that dynamical criticality in genetic networks naturally emerges as a consequence of a delicate balance between two fundamental forces in evolution: the conservation of already acquired phenotypes (phenotypic robustness) and the generation of new ones (phenotypic innovation). We do so by evolving populations of random Boolean networks that can mutate and grow via gene duplication, and that are subjected to a Darwinian selection process that focuses on the conservation and innovation of the dynamical attractor landscape of the network population. Our results show that the trade-off between robustness and innovation suffices to make all the networks in the population evolve towards criticality. Furthermore, preliminary results indicate that the existence or absence of hubs (global regulators) similar to the ones observed in real genetic networks, is determined by the information content of the attractors through the evolutionary process.

President Obama’s May 2011 policy statement on "International Strategy for Cyberspace: Prosperity, Security, and Openness in a Networked World," lays out many of the principles for preventing harm and realizing the potential of this revolutionary technology. Yet, specific policy measures to ensure adequate cyber defenses in the private sector are not yet in place. Furthermore, the 2011 document fails to address adequately the uses of cyber operations by the United States in attacks on other countries' critical infrastructure. Revelations about the US deployment of malware in a cyber attack on Iran's nuclear centrifuges in 2009, and increased attention to the perils of damaging attacks on US financial, military, and infrastructure operating systems suggest the need for robust public policy recommendations and implementation. This talk will examine underlying assumptions and frameworks for policy development.

Certain freshwater fish from South America and Africa are able to sense their surroundings by emitting weak, millivolt-level electric discharges and analyzing the perturbations arising from nearby objects. This ability allows them to hunt, navigate, and communicate at night and in turbid waters. Weakly electric fish are able to localize targets in 3D space, assess target characteristics such as size, shape and electrical impedance, and recognize the sex, social status and perhaps individual identities of other fish. This talk will provide an overview of some of the key neural mechanisms and computational strategies involved in carrying out these challenging information processing tasks.

It is widely believed that synaptic modifications underlie learning and memory. This hypothesis has led to the study of many `learning rules' that implement in a simplified way how synaptic efficacy is controlled by neuronal activity. This talk will focus on a complementary research direction: investigating optimal storage properties in neural circuits. The first part of the talk will focus on the perceptron, the simplest feed-forward network model, as a simplified model of the granule-Purkinje cell pathway in the cerebellum. The distribution of synaptic weights of a perceptron that optimizes storage capacity can be computed using methods from statistical physics. This distribution has two striking features: (i) it contains a large number (at least 50%) of exactly zero weights (`silent' or `potential' synapses); (ii) positive weights are distributed according to a monotonically decreasing function. We find that the theoretical distribution fits closely the distribution of synaptic weights of connections between granule cells and Purkinje cells, suggesting Purkinje cells function close to their optimal capacity in adult rats, which we estimate to be about 5Kb per cell. In the second part of the talk, I will consider a network with a fully connected recurrent architecture, as a simplified model for local pyramidal cell networks of neocortex. In networks storing a large number of fixed point attractor states, the distribution of synaptic weights turns out to be exactly the same as the one for a perceptron, and hence contains a large fraction of `silent', or `potential' synapses. Finally, I will consider other statistical properties of the connectivity matrix (e.g.the joint distribution of synaptic weights for pairs of neurons), and compare the theoretical results with recently published data on synaptic connectivity in cortical slices.

Our Universe has a finite observable volume, and therefore within our Universe there is a unique most massive object. This object will be a cluster of galaxies. Computational and theoretical studies of the growth of structure have matured, and the mass of the most massive objects can now be robustly predicted to the level of a few percent. Furthermore, it is now possible to observe volume-limited samples of high-mass clusters. If objects are found with excessively large masses, or insufficient objects are found near the maximum expected mass, this would be a strong indication of the failure of our standard cosmological model. We show that preliminary observations are roughly consistent with expectations, with a few suggestive outliers.

At a time when microbial ecology is largely traveling along genomic roads, we cannot forget that the functions and services of microbes depend greatly on their behaviors, encounters, and interactions with their environment. New technologies, including microfluidics and high-speed video microscopy, provide a powerful opportunity to spy on the lives of microbes, directly observing their behaviors at the spatiotemporal resolution most relevant to their ecology, and enabling a deeper understanding of the biophysical mechanisms underpinning these behaviors. I will illustrate this 'quantitative natural history approach' to microbial ecology by focusing on marine bacteria, unveiling striking adaptations in their motility and chemotaxis and describing how these are connected to their incredibly dynamic, gradient-rich microenvironments. Specifically, I will present (i) sub-micrometer imaging of single cells at up to thousand frames per second, demonstrating that marine bacteria have a unique mode of swimming, exploiting a mechanical buckling instability of their flagellum to reorient; and (ii) microfluidic experiments that capture the dramatic chemotactic abilities of marine bacteria, including bacterial pathogens storming towards the roiling surface of their coral hosts. Through these examples, I aim to illustrate how we can use direct visualization to learn about the biophysical mechanisms and the ecological implications of the behaviors of the smallest of life forms.

I will begin by discussing experiments in which we observe the emergence of localized, crumpled features from smoothly deformed sheets. I will then move on to describe the 3-dimensional spatial structure of highly crushed elastic and plastic sheets. One of the objectives of our experiments is to understand the effect of the protocol and boundaries on the crumpling process. We follow the development of stress-focused structures as well as the emergent organization of structural elements, such as the condensation of planar facets. I will end by discussing the evidence for an ordered state underlying the complex, seemingly-disordered geometry of a crumpled object.

By decreasing the temperature the relaxation time of super-cooled liquids increases by more than ten orders of magnitude. This feature, which is the hallmark of the glass transition, is also the main obstacle to study it because liquids inevitably fall out of equilibrium before showing a genuine critical behavior. In this talk I propose a way to short-circuit this problem, cross the phase transition and sample the ideal glass in equilibrium. It is based on the idea of pinning particles at random from an equilibrium configuration. Contrary to cooling protocols, immediately after pinning the remaining free particles are automatically in equilibrium. Thus, the ideal glass can be reached easily: one has just to pin enough particles. This makes possible a whole new set of investigations, in particular studying the glass transition approaching it from both sides. I will present the theory of the glass transitions induced by random pinning, discuss generalizations of the pinning procedure and differences with other approaches recently presented in the literature.

Dynamical systems have a natural tendency to relax through dissipative processes to a minimum-energy state, subject to any relevant constraints. An example is provided by the relaxation of a magnetic field in a perfectly conducting but viscous fluid, subject to the constraint that the magnetic field lines are frozen in the fluid. One may infer the existence of magnetostatic equilibria (and analogous steady Euler flows) of arbitrary field-line topology. In general, discontinuities (current sheets) appear during this relaxation process, and this is where reconnection of field-lines (with associated change of topology) can occur. Slow change of boundary conditions can lead to critical conditions in which such topological jumps are inevitable.

A simple example of this type of behaviour that can be realised in the laboratory is provided by a soap-film bounded by a flexible wire (or wires) which can be continuously and slowly deformed. At each instant the soap-film is relaxed in quasi-static manner to a minimum-area (i.e. minimum-energy) state compatible with the boundary configuration. This can however pass through a critical configuration at which a topological jump is inevitable. We have studied an interesting example of this behaviour: the jump of a one-sided (Möbius strip) soap-film to a two-sided film as the boundary is unfolded and untwisted from the double cover of a circle. The nature of this jump will be demonstrated and explained.

Random Matrix Theory was initially developed to explain the eigen-energy distribution of heavy nuclei. It has become clear by now that its domain of application is much broader and extends to very different fields such as number theory and quantum chaos, just to cite a few. In particular, it has been conjectured—and proved or verified in some special cases—that quantum ergodic (or chaotic) systems are characterized by eigen-energies statistics in the same universality class of random matrices and by eigen-functions that are delocalized over the configuration space. On the contrary, non-ergodic quantum systems, such as integrable models, are expected to display a Poisson statistics of energy levels and localized wave-functions. Starting from Anderson’s pioneering papers, similar properties have also been studied for electrons hopping in a disordered environment. Remarkably, also in this case, similar features of the energy-level statistics have been found. All that has lead to the conjecture that delocalization in configuration space, ergodicity and level statistics are intertwined properties.

In this talk we revisit the old problem of non-interacting electrons hopping on a Bethe lattice with on-site disorder. By using numerical simulations, the cavity method and mapping to directed polymers in random media we unveil the existence of an intermediate phase in which wave-functions are delocalized but the energy-level statistics is Poisson. This new phase, in which the system is non-ergodic but delocalized, may play an important role in several fields from random matrix theory to strongly interacting quantum disordered systems, in particular it could be related to the non-ergodic metallic phase conjectured to exist in the context of Many-Body Localization.

We generate nematic droplets with handles and stabilize them against surface-tension instabilities using yield-stress fluids. For toroidal droplets, the nematic spontaneously twists; this happens for all, slender and fat tori, as a result of saddle-splay contributions to the elastic free energy. The addition of handles is accompanied by the presence of defects in the order. We find there are two -1 defects per handle, located in regions with saddle geometry.

Properties of matter not adequately described by classical physics have gained a lot of attention in quantum information science and condensed matter physics. The non-classcality is mostly attributed to "entanglement", which can be utilized for quantum computing and yet makes the study of quantum matter on classical computers so difficult.

In this talk we first discuss the ground state properties of quantum spin chains and trees with generic local interactions within the framework of Matrix Product States formulation. We then present the first example of a Frustration Free translation-invariant spin-1 chain that has a unique highly entangled ground state and exhibits some signatures of a critical behavior. The mathematical techniques used here may be of independent interest.

The process of diffusion is a consequence of particle number conservation and locality, in systems with sufficient damping. Diffusive processes are ubiquitous in nature — familiar examples range from pollen grains executing Brownian motion in water, to electrical charge transport through a power line. Yet fundamental questions remain to be addressed in the quantum mechanical formalisms leading to diffusion from a microscopic theory. In this talk I shall first elucidate the standard field theoretical technique for deriving the law of diffusion for electrons in a crystal, in the presence of disorder. Through this example, I shall also introduce our new analytical tools for addressing anisotropic scattering, thus deriving the Boltzmann diffusion constant via the diagrammatic approach. Next I shall introduce Weyl semimetals, an exotic material with topologically protected gapless surface states with discontinuous Fermi surfaces, arising from the valence and conduction bands touching in three dimensions. The derivation of the correct diffusion law for this material using the previously introduced techniques will then be outlined. I shall conclude by showing how this is a diffusive process with a nontrivial memory function and discussing the experimental consequences of this result.

Flow through nanometer-scale pores is important in biology, geological flows through porous rocks, and in filtration and separation processes. Flow through these pores is controlled by different physics than conventional macroscopic pipes. Single phase fluid flow in nanometer channels is extremely sensitive to slip and the exact boundary conditions at the solid- liquid interface. Because curvature of interfaces in small channels is very high, two phase flows are subject to large Laplace pressures and are sensitive to the contact angle and the wetting properties of the interface. Experiments by others on flow through “carpets” of carbon nanotubes have yielded flow rates that are hundreds of times greater than predicted by conventional hydrodynamics. I will discuss recent experiments in our lab on pressure driven flow through single nanopipes using water, nitrogen and helium (including superfluid) over a wide range of temperature. The results show clear transitions between distinct flow regimes. I will also discuss contact angle measurements of water on graphite. Somewhat paradoxically, water does not wet many materials. Our experiments have identified the first wetting transition in water at T=280C.

Drop coalescence has been thoroughly studied, especially for two liquid drops coalescing in vacuum or air. However, little is known about how the surrounding fluid influences the process when the two drops are surrounded by an external fluid with significant density or dynamic viscosity. This two-fluid coalescence is the important situation that appears commonly in nature and industry. We use a combination of high-speed imaging and an ultrafast electrical method to study coalescence in this regime. We find that even if the outer fluid is 50 times more viscous than the drops, the coalescence speed need not be affected, even at the earliest moments. To understand the behavior of the outer fluid in isolation, we turn to air bubbles coalescing inside a very viscous external liquid. Our data is consistent with a simple ansatz that the length-scale for the flows is much larger in the outer fluid than in the drops, so that viscous stresses in the drops will always dominate at the beginning. This leaves us with the parting lesson: “it’s what’s inside that counts”.

While the concept of evolution as mechanism for iterative improvement has existed for centuries, three key developments in the past 20 years have made it promising framework for science and engineering in the 21st century. First, artificial variants of evolution, specifically those implemented through computer algorithms, have successfully transformed the biological concept into a robust and efficacious optimization method. Second, computer simulations of a variety of physical systems are now fast enough to be effectively used to compare the relative quality of a potential design against a small population of alternatives. Third, advances in rapid prototyping, such as 3D printing, present a means to physically realize complicated or unconventional designs that may result from unconstrained evolution in a digital environment.

In this talk, I will discuss examples of how these three developments may be synthesized to address particular problems in granular physics. First, I will show how artificial evolution can be used with simulations and 3D printing to explore the role of particle shape in granular materials. This approach can be used to discover new, unexpected particles that form the stiffest packings, softest packings, and even exotic configurations that stiffen, rather than weaken, under compression. Furthermore, experiments on 3D printed versions of digitally designed particles directly validate these results.

Outside of mechanical response, I will present results about shapes that pack densest when poured into a container. In this case, I will show how artificial evolution can be used to not only identify particularly optimal shapes, but also to explore the role of physical variables, like friction, or the packing procedure itself.

Finally, I will show a variation on these themes that performs evolution directly in the lab by leveraging the jamming concept to rapidly prototype experiments. In this case, a fully automated experiment adjusts whether patches of material are liquid-like or solid-like to discover configurations that, when loaded at one side, transmit the lowest force to the other. In spite of a search space of roughly 2^36 candidates, evolutionary algorithms are able to find a particularly novel solution in just under 12 hours.

In the 1970’s our low temperature research group was drawn into a new line of research concerning the odor from a Springfield paper plant operated by the Weyerhauser company. It occurred to me that passing the flue gases through a cold trap would likely solve the problem. We sought out the management, who were quite receptive to the idea of a cold trap, and supported research which proved to be quite effective. Based on this experience, we applied to NSF for a study of trapping SO2 from coal plants and published a paper on the laboratory results. We also engaged the Bechtel Power Company to do an engineering study of the possibilities for a major coal plant. One problem we encountered was the simultaneous production of CO2, which at the time was not considered to be as important as it is today. This talk will discuss the possibilities, thermodynamics, and added costs of capturing CO2 cryogenically, as an example of how physics can be applied to urgent world-wide problems.

A drop impacting a solid surface with sufficient velocity will splash and emit many small droplets. Surprisingly, however, removing the ambient air suppresses splashing completely. There are several distinct regimes of splashing that all display this extreme sensitivity to gas pressure. In particular, decreasing the air pressure also eliminates splashing on a rough surface even though the overall form of the splash is significantly different in this regime. The mechanism underlying how the surrounding gas affects splashing remains unknown. Our approach to unveil this mystery is twofold. (i) We establish a unique criterion that predicts the transition from smooth deposition to splashing, taking into account all the relevant liquid and gas properties. (ii) We search for where on the drop surface the air matters. We look closely below the drop, at the air flow above the drop and at the moving contact line near the drop edge.

What can one uncover by simply watching a bacterial cell grow and divide, over and over again? It turns out, a lot! Evidently, one learns how the size of each cell increases with time. More interestingly, one finds the nature of the nuanced interplay between biological length scales and time scales, which determines when the size of a cell is just right for it to divide. Further, one could even discover that there is but one primary time scale, a cellular ``unit of time'', which governs the nature of fluctuations in both growth and division, and thus make nontrivial predictions for how these fluctuations must scale when ambient conditions are changed. Finally, one could ask how all of this is revised when the cell is forced to senesce (``age'') by being subjected to inclement nutrient conditions. In this talk I shall first introduce the experimental technology that we have developed, which allows for probing such issues, and then discuss some of our surprising findings.

The dynamics of how two rough frictional interfaces detach is a fundamental question in fields ranging from material science to geophysics. On the one hand, the onset of frictional motion is thought to be characterized by the static friction coefficient that couples two materials. For hundreds of years, this has been considered to be a material constant. On the other hand, the same processes that give rise to the onset of frictional motion also cause earthquakes, when tectonic plates locked together by friction start to slip. The frictional interface that locks two macroscopic blocks of material together composed of an ensemble of discrete microscopic contacts that give the interface its strength. We will present new experiments that show how the onset of frictional motion is caused by the fracture of these contacts. This takes place via rapid earthquake-like rupture processes that immediately precede the onset of macroscopic motion that we know as frictional sliding. We will show that both a number of different modes of earthquakes exist and that the “static friction coefficient” is not a material constant at all, but is intimately related to the details of how forces are applied to a system. We will present new results that describe the detailed spatio-temporal processes that take place within a rough and dry frictional interface.

Over the last 15 years glaciers that discharge directly into the ocean have shown pronounced increases in flow speed, thinning, and retreat. The resulting increased discharge of ice contributes to an increased rate of sea level rise; in Greenland this has put the ice sheet tens of percent out of balance. While the changes in these tidewater outlet glaciers are large, understanding their direct cause remains elusive. I will present observations of glacier-ocean interaction that range from half-gigaton calving events to tidal and seasonal variations in flow speed from radar interferometry and timelapse photography. In addition, I will summarize recent work on the relationship between fjord circulation and melting on an Alaskan tidewater glacier, to introduce the problem of energy exchange between warming sea water and a rapidly flowing glacier, and to highlight a little appreciated tie between increased surface melt on the ice and the stability of a glacier terminus.

Delay match to sample experiments (DMS) involve the presentation of a cue stimulus, several distractor stimuli and finally a test stimulus that is either the same as the cue or not. These experiments have been used for decades to study how the brain develops long term memory,and then retains in working memory information about the cue. The electrophysiology of DMS experiments has inspired much of the modeling in attractor networks, in particular the observation of persistent activity of neurons selective to learned objects during the delay period prior to the test stage. The basic paradigm posits that stimulation with learned patterns leads to sustained activity due to the learned recurrent connections in the network. I will introduce the recurrent network in its simplest version - binary neurons and binary synapses, and present various results on memory capacity in this framework; how many stimuli can such a network learn and subsequently retrieve and sustain in working memory. I will then raise several questions that are rarely addressed in the literature. How does the network ensure that the cue stays in memory? How does the network figure out that the test pattern is or is not active in memory? How does the network avoid distractor repetitions? Additional interesting phenomena have been observed in more recent experiments on repetition detection - delay match to multiple sample tasks (DMMS) - where the cue could be any of the patterns in the sequence. Here it is observed that performance is better with novel patterns than with learned patterns. I will show how all these phenomena can be handled within the simple attractor network framework using perturbations of certain global parameters such as inhibition, noise level and depression rate.

When handling a large sheet of paper one must take special care not to impose boundary conditions which result in local strains that yield the material and leave permanent defects. Previous studies described paper creases as loci of bending deformations that minimize the elastic energy of the confined sheet. In the elastic framework elongated creases result from the yielding of ridges formed by the elastic interaction between two point-like singular structures termed d-cones. I will describe our experimental study of a new type of crease that is inherently plastic and forms by the propagation of a single point-defect. We strain a preexisting d-cone in an elsto-plastic thin sheet and show it remains pinned up to a critical loading force, Fy, which scales quadratically with the thickness of the thin sheet. When Fy is reached, the singular structure at the apex of the d-cone sharpens abruptly. The resulting focusing of strains yields the material just ahead of the d-cone, allowing it to propagate, leaving a furrow-like scar in its wake.

Robert Almgren will talk about his scientific and professional evolution: from MRSEC to the wilds of Wall Street, from tenured professor, to managing director at an investment bank, to entrepreneur. Despite the changing outward context, similar scientific habits and tools are useful in all these settings. He will describe the scientific challenges he encountered along the way and will emphasize the common elements between microstructure of fluids and microstructure of financial markets.

We show that the quantum Hall states exhibit a special type of symmetry, which is best captured by the formalism of Newton-Cartan gravity. We derive various physical implications of the symmetry, including the wavenumber dependence of the Hall conductivity, various sum rules and inequalities. The Laughlin wavefunction is found to saturate one of the inequalities. We present a model where the magneto-roton is interpreted as a emergent massive graviton, with consequences for experiments.

We present a mathematical perspective on two examples of numerical simulations in materials science that require a dedicated approach because the problems are multiscale in nature. The first example is the Parallel Replica dynamics, useful in computational statistical Physics to directly and accurately simulate technologically relevant systems such as the plastic deformation of carbon nanotubes over timescales that can reach milliseconds on current state-of-the-art petascale computing platforms. The aim of the approach is to efficiently generate a coarse-grained evolution of a given stochastic process. The mathematical study allows to theoretically assess the performance of the approach and improve it. The second example is related to the modeling of some composite materials in use in the aerospace industry. The simulation of the mechanical response of such fiber reinforced materials is a computational challenge. Several space scales are relevant and the distribution of the microstructures is highly heterogeneous. Specific numerical approaches, in the spirit of the homogenization method, successfully address the problem. Their mathematical study again provide useful information on how to optimize their efficiency and leverage the power of parallel computers.

Why and how we sleep has been a matter of speculation and study for millennia. Every day our brains cycle between waking and sleeping states. Both of these brain states are highly active, but the nature of the activity and the connection of the brain to the outside world in each state are distinct. Primate and rodent model systems have provided great insights into sleep, but the circuitry in these organisms is quite complex. The recent finding that insects sleep suggests that Drosophila melanogaster, a simple and genetically tractable organism, can be used to study this process.

In recent years, work from my lab and others has exploited the new genetic and electrophysiological tools available in Drosophila to push forward our understanding of sleep by identification and manipulation of the underlying circuitry. In this talk I will discuss the evolutionary conservation of sleep at the behavioral and circuit levels in the fly and how dissection of the circuitry in this organism may allow us to understand the fundamental nature of sleep regulation.

Combining knowledge of physics with a healthy willingness to estimate and approximate, we can say some rather profound things about future paths available to our society in terms of energy and resources. Topics such as growth, global warming, fossil fuels and their potential replacements, and energy storage are ripe targets for back-of-the-envelope quantification, and will be explored in this talk. A subtext is that we should not take for granted that superior substitutes will replace fossil fuels.

Long-range quantum order underlies a number of related physical phenomena including superfluidity, superconductivity, the Higgs mechanism, Bose-Einstein condensates, and spin systems. While superfluidity in Helium-4 was one of the earliest discovered of these, it is not the best understood, owing to locally-strong interactions which make theoretical progress difficult, and a lack of local experimental probes. Our group discovered that micron-sized hydrogen particles, and most recently fluorescent nanoparticles, may be used to label quantized vortices in flows of superfluid helium. Particles not on vortices trace the motion of the normal component of the superfluid. This diagnostic tool has given us a new perspective on an old subject. By directly observing and tracking these particles, we have confirmed the two-fluid model, observed vortex rings and reconnection, characterized thermal counterflows, observed Kelvin waves, and taken local observations of the very peculiar nature of quantum turbulence. One of many surprising observations is the existence of power law tails in the probability distribution of velocity for these flows. That was predicted by our group and verified as stemming from the reconnection of quantized vortices. We conclude that quantum turbulence is dominated by reconnection and vortex ring collapse, making turbulence in a quantum liquid distinct from classical turbulence of a Newtonian fluid.

Progress in atomic physics and quantum information science has motivated much recent study of the behavior of strongly-interacting many-body quantum systems fully isolated from their environment, and thus undergoing coherent quantum time evolution. What does it mean for such a system to go to thermal equilibrium? I will explain the Eigenstate Thermalization Hypothesis (ETH), which says that each individual exact eigenstate of the system's Hamiltonian is at thermal equilibrium, and which appears to be true for most (but not all) quantum many-body systems. Prominent among the systems that do not obey this hypothesis are quantum systems that are many-body Anderson localized. These "many-body localized" systems can retain local memory of local properties of their initial state for infinite time, and thus do not thermally equilibrate and are of interest for quantum information storage. A key issue here is whether or not the system itself constitutes a "reservoir" that can equilibrate its parts.