Entropy, Causal Diagrams and Warped Flat Spacetimes
Raphaela Wutte Technische Universität Wien
4093 - 4104 of 19185 Results
Format results
Quantum many-body topology of crystals and quasicrystals
Dominic Else Perimeter Institute for Theoretical Physics
Predicting many properties of quantum systems with chaotic dynamics
Hong-Ye Hu University of California, San Diego
Representation Costs of Linear Neural Networks: Analysis and Design
Mina Karzand (University of California, Davis)Is Overfitting Actually Benign? On the Consistency of Interpolating Methods
Preetum Nakkiran (UCSD)The Spectrum of Nonlinear Random Matrices for Ultra-Wide Neural Networks
Yizhe Zhu (University of California, Irvine)Exact Asymptotics and Universality for Gradient Flows and Empirical Risk Minimizers
Andrea Montanari (Stanford University)Paleo-Detectors - Digging for Dark Matter and Neutrinos
Sebastian Baum Nordic Institute for Theoretical Physics
Learning Staircases
Emmanuel Abbe (École polytechnique fédérale de Lausanne), Enric Boix (MIT), Theodor Misiakiewicz (Stanford University)
Entropy, Causal Diagrams and Warped Flat Spacetimes
Raphaela Wutte Technische Universität Wien
Conformal field theories have played an important role in understanding the origin of gravitational entropy, as the Cardy formula accounts for the entropy of three-dimensional BTZ black holes or black holes with an AdS3 near-horizon region. More recently, Cardy-type formulas have also been derived for different field theories, such as warped conformal field theories, which are two-dimensional field theories invariant under chiral scaling symmetry.
In this talk I discuss a particular class of three-dimensional spacetimes, dubbed warped flat spaces. I will present their geometric and thermodynamic properties, focusing on their causal structure. I will explain how their entropy can be accounted for through a dual description in terms of a warped conformal field theory.
Quantum many-body topology of crystals and quasicrystals
Dominic Else Perimeter Institute for Theoretical Physics
When an interacting quantum many-body system is cooled down to its ground state, there can be discrete "topological invariants" that characterize the properties of such ground states. This leads to the concept of "topological phases of matter" distinguished by these topological invariants. Experimental manifestations of these topological phases of matter include the integer and fractional quantum Hall effect, as well as topological insulators.
In this talk, after a general overview of topological phases of matter, I will explain how to define topological invariants that are specific to the ground states of regular crystals, i.e. systems that are periodic in space. I will discuss the physical manifestations of the resulting "crystalline topological phases", including implications for the properties of crystalline defects such as dislocations and disclinations. Then, I will explain how these ideas can be generalized to quasicrystals, which are a different class of materials that have long-range spatial order without exact periodicity. These ideas ultimately lead to a general classification principle for crystalline and quasicrystalline topological phases of matter.
Predicting many properties of quantum systems with chaotic dynamics
Hong-Ye Hu University of California, San Diego
Classical shadow tomography provides an efficient method for predicting functions of an unknown quantum state from a few measurements of the state. It relies on a unitary channel that efficiently scrambles the quantum information of the state to the measurement basis. However, it is quite challenging to realize deep unitary circuits on near-term quantum devices, and an unbiased reconstruction map is non-trivial to find for arbitrary random unitary ensembles. In this talk, I will discuss our recent progress on combining classical shadow tomography with quantum chaotic dynamics. Particularly, I will introduce two new families of shadow tomography schemes: 1) Hamiltonian-driven shadow tomography and 2) Classical shadow tomography with locally scrambled quantum dynamics. In both works, I’ll derive the unbiased reconstruction map, and analyze the sample complexity. In the Hamiltonian-driven scheme, I will illustrate how to use proper time windows to achieve a more efficient tomography. In the second work, I will demonstrate advantages of shadow tomography in the shallow circuit region. Then I’ll conclude by discussing approximate shadow tomography with local Hamiltonian dynamics, and demonstrate that a single quench-disordered quantum spin chain can be used for approximate shadow tomography.
References:
[1] Hong-Ye Hu, Yi-Zhuang You. “Hamiltonian-Driven Shadow Tomography of Quantum States”. arXiv:2102.10132 (2021)
[2] Hong-Ye Hu, Soonwon Choi, Yi-Zhuang You. “Classical Shadow Tomography with Locally Scrambled Quantum Dynamics”. arXiv: 2107.04817 (2021)Zoom Link: https://pitp.zoom.us/j/99011187936?pwd=OVU3VkpyZ21YcXRCOW5DOHlnSWlVQT09
Representation Costs of Linear Neural Networks: Analysis and Design
Mina Karzand (University of California, Davis)For different parameterizations (mappings from parameters to predictors), we study the regularization cost in predictor space induced by l_2 regularization on the parameters (weights). We focus on linear neural networks as parameterizations of linear predictors and identify the representation cost of certain sparse linear ConvNets and residual networks. In order to get a better understanding of how the architecture and parameterization affect the representation cost, we also study the reverse problem, identifying which regularizers on linear predictors (e.g., l_p quasi-norms, group quasi-norms, the k-support-norm, elastic net) can be the representation cost induced by simple l_2 regularization, and designing the parameterizations that do so.On the Cryptographic Hardness of Learning One-Hidden Layer Neural Networks
Ilias Zadik (MIT)In this short talk, I will share some recent progress on the hardness of learning shallow RELU neural networks (Relu-NN) and polynomially small adversarial noise. We will present a result that efficiently learning an 1-hidden layer Relu-NN under Gaussian input and adversarial noise is "cryptographically hard", in the sense that it implies a polynomial-time quantum algorithm for the worst-case shortest vector problem.Is Overfitting Actually Benign? On the Consistency of Interpolating Methods
Preetum Nakkiran (UCSD)This talk will be very informal: I will discuss ongoing research with open questions and partial results. We study the asymptotic consistency of modern interpolating methods, including deep networks, in both classification and regression settings. That is, we consider learning methods which scale the data size while simultaneously scaling the model size, in a way which always interpolates the train set. We present empirical evidence that, perhaps contrary to intuitions in theory, many natural interpolating learning methods are *inconsistent* for a wide variety of distributions. That is, they do not approach Bayes-optimality even in the limit of infinite data. The message is, in settings with nonzero Bayes risk, overfitting is not benign: interpolating the noise significantly harms the classifier, to the point of preventing consistency. This work is motivated by: (1) understanding differences between the overparameterized and underparameterized regime, (2) guiding theory towards more "realistic" assumptions to capture deep learning practice, and (3) understanding common structure shared by "natural" interpolating methods. Based on joint work with: Neil Mallinar, Amirhesam Abedsoltan, Gil Kur, and Misha Belkin. And is a follow-up work to the paper "Distributional Generalization", joint with Yamini Bansal.
Subdiffusion and ergodicity breaking in systems with emergent or microscopic dipole-moment conservation
Alan Morningstar Citadel LLC
I will first give a brief overview of my research in the field of out-of-equilibrium quantum many-
body physics, ranging from the theory of many-body localization, to the recent application of TensorProcessing Units for accelerating simulations of quantum dynamics. I’ll then focus on (1) the
experimental observation and theoretical explanation of subdiffusive dynamics in a “tilted” Fermi-
Hubbard system [PRX 10, 011042 (2020)], and (2) a “freezing” phase transition between weak andstrong ergodicity breaking in systems with particles that are immobile by themselves, but undergo
coordinated pair hopping [PRB 101, 214205 (2020)]. These topics contain the common thread of
either an emergent or microscopic conservation of the dipole moment (center of mass of the particle
distribution), and I will provide simple pictures for how this leads to the subdiffusion and ergodicity
breaking.Universality of Neural Networks
Dan Mikulincer (MIT)It is well known that, at a random initialization, as their width approaches infinity, neural networks can be well approximated by Gaussian processes. We quantify this phenomenon by providing non-asymptotic convergence rates in the space of continuous functions. In the process, we study the Central Limit Theorem in high and infinite dimensions, as well as anti-concentration properties of polynomials with random variablesThe Spectrum of Nonlinear Random Matrices for Ultra-Wide Neural Networks
Yizhe Zhu (University of California, Irvine)We obtain limiting spectral distribution of empirical conjugate kernel and neural tangent kernel matrices for two-layer neural networks with deterministic data and random weights. When the width of the network grows faster than the size of the dataset, a deformed semicircle law appears. In this regime, we also calculate the asymptotic test and training errors for random feature regression. Joint work with Zhichao Wang (UCSD).Exact Asymptotics and Universality for Gradient Flows and Empirical Risk Minimizers
Andrea Montanari (Stanford University)We consider a class of supervised learning problems whereby we are given n data points (y_i,x_i), with x_i a d-dimensional dfeature factor, y_i a response, and the model is parametrized by a vector of dimension kd. We consider the high-dimensional asymptotics in which n,d diverge, with n/d and k of order one. As a special case, this class of models includes neural networks with k hidden neurons. I will present two sets of results: 1. Universality of certain properties of empirical risk minimizers with respect to the distribution of the feature vectors x_i. 2. A sharp asymptotic characterization of gradient flow in terms of a one-dimensional stochastic process. [Based on joint work with Michael Celentano, Chen Cheng, Basil Saeed]Paleo-Detectors - Digging for Dark Matter and Neutrinos
Sebastian Baum Nordic Institute for Theoretical Physics
Paleo-Detectors are natural minerals which record damage tracks from nuclear recoils over geological timescales. Minerals commonly found on Earth are as old as a billion years, and modern microscopy techniques may allow to reconstruct damage tracks with nanometer scale spatial resolution. Thus, paleo-detectors would constitute a technique to achieve keV recoil energy threshold with exposures comparable to a kiloton-scale conventional "real-time" detector. In this talk, I will discuss the potential of paleo-detectors for the direct detection of dark matter as well as for detecting low-energy neutrinos as are e.g. emitted by core collapse supernovae or our Sun. Furthermore, the age of the minerals provides the ability to look back across gigayear-timescales, giving paleo detectors the unique ability to probe changes in the cosmic ray rate or the galactic supernova rate over such timescales as well as dark matter substructure Earth might have encountered during its past few trips around our Galaxy.
Learning Staircases
Emmanuel Abbe (École polytechnique fédérale de Lausanne), Enric Boix (MIT), Theodor Misiakiewicz (Stanford University)It is known that arbitrary poly-size neural networks trained by GD/SGD can learn in class in SQ/PAC. This is however not expected to hold for more regular architectures and initializations. Recently, the staircase property emerged as a condition that seems both necessary and sufficient for certain regular networks to learn with high accuracy, with the positive result established for sparse homogeneous initializations. In this talk, we show that standard two-layer architectures can also learn staircases with features being learned over time. It is also shown that kernels cannot learn staircases of growing degree. Joint work with Enric Boix-Adsera and Theodor Misiakiewicz.