-
Extinctions as a vestige of instability: the geometry of stability and feasibility
Authors:
Stav Marcus,
Ari M. Turner,
Guy Bunin
Abstract:
Species coexistence is a complex, multifaceted problem. At an equilibrium, coexistence requires two conditions: stability under small perturbations; and feasibility, meaning all species abundances are positive. Which of these two conditions is more restrictive has been debated for many years, with many works focusing on statistical arguments for systems with many species. Within the framework of t…
▽ More
Species coexistence is a complex, multifaceted problem. At an equilibrium, coexistence requires two conditions: stability under small perturbations; and feasibility, meaning all species abundances are positive. Which of these two conditions is more restrictive has been debated for many years, with many works focusing on statistical arguments for systems with many species. Within the framework of the Lotka-Volterra equations, we examine the geometry of the region of coexistence in the space of interaction strengths, for symmetric competitive interactions and any finite number of species. We consider what happens when starting at a point within the coexistence region, and changing the interaction strengths continuously until one of the two conditions breaks. We find that coexistence generically breaks through the loss of feasibility, as the abundance of one species reaches zero. An exception to this rule - where stability breaks before feasibility - happens only at isolated points, or more generally on a lower dimensional subset of the boundary.
The reason behind this is that as a stability boundary is approached, some of the abundances generally diverge towards minus infinity, and so go extinct at some earlier point, breaking the feasibility condition first. These results define a new sense in which feasibility is a more restrictive condition than stability, and show that these two requirements are closely interrelated. We then show how our results affect the changes in the set of coexisting species when interaction strengths are changed: a system of coexisting species loses a species by its abundance continuously going to zero, and this new fixed point is unique. As parameters are further changed, multiple alternative equilibria may be found. Finally, we discuss the extent to which our results apply to asymmetric interactions.
△ Less
Submitted 18 May, 2024;
originally announced May 2024.
-
Motion from Measurement: The Role of Symmetry of Quantum Measurements
Authors:
Luka Antonic,
Yariv Kafri,
Daniel Podolsky,
Ari M. Turner
Abstract:
In quantum mechanics, measurements are dynamical processes and thus they should be capable of inducing currents. The symmetries of the Hamiltonian and measurement operator provide an organizing principle for understanding the conditions for such currents to emerge. The central role is played by the inversion and time-reversal symmetries. We classify the distinct behaviors that emerge from single a…
▽ More
In quantum mechanics, measurements are dynamical processes and thus they should be capable of inducing currents. The symmetries of the Hamiltonian and measurement operator provide an organizing principle for understanding the conditions for such currents to emerge. The central role is played by the inversion and time-reversal symmetries. We classify the distinct behaviors that emerge from single and repeated measurements, with and without coupling to a dissipative bath. While the breaking of inversion symmetry alone is sufficient to generate currents through measurements, the breaking of time-reversal symmetry by the measurement operator leads to a dramatic increase in the magnitude of the currents. We consider the dependence on the measurement rate and find that the current is non-monotonic. Furthermore, nondegenerate measurements can lead to current loops within the steady state even in the Zeno limit.
△ Less
Submitted 9 May, 2024;
originally announced May 2024.
-
Unvortex Lattice and Topological Defects in Rigidly Rotating Multicomponent Superfluids
Authors:
Roy Rabaglia,
Ryan Barnett,
Ari M. Turner
Abstract:
By examining the characteristics of a rotating ferromagnetic spinor condensate through the perspective of large spin, we uncover a novel kind of topological point defect in the magnetization texture. These defects are predicted not by the conventional homotopy group analysis but by the Riemann-Hurwitz formula. The spin texture in the system is described by an equal-area map** from the plane to t…
▽ More
By examining the characteristics of a rotating ferromagnetic spinor condensate through the perspective of large spin, we uncover a novel kind of topological point defect in the magnetization texture. These defects are predicted not by the conventional homotopy group analysis but by the Riemann-Hurwitz formula. The spin texture in the system is described by an equal-area map** from the plane to the sphere of magnetization, forming a lattice of uniformly charged Skyrmions. This lattice carries doubly-quantized (winding number = 2) point defects arranged on the sphere in a tetrahedral configuration. The fluid is found to be rotating rigidly, except at the point defects where the vorticity vanishes. This vorticity structure contrasts with the well-known vortex lattice in scalar rotating superfluids, where vorticity concentrates exclusively within defect points, forming an unconventional "unvortex" lattice. Numerical results are presented, which are in agreement with the aforementioned predictions.
△ Less
Submitted 8 April, 2024;
originally announced April 2024.
-
Steering Llama 2 via Contrastive Activation Addition
Authors:
Nina Panickssery,
Nick Gabrieli,
Julian Schulz,
Meg Tong,
Evan Hubinger,
Alexander Matt Turner
Abstract:
We introduce Contrastive Activation Addition (CAA), an innovative method for steering language models by modifying their activations during forward passes. CAA computes "steering vectors" by averaging the difference in residual stream activations between pairs of positive and negative examples of a particular behavior, such as factual versus hallucinatory responses. During inference, these steerin…
▽ More
We introduce Contrastive Activation Addition (CAA), an innovative method for steering language models by modifying their activations during forward passes. CAA computes "steering vectors" by averaging the difference in residual stream activations between pairs of positive and negative examples of a particular behavior, such as factual versus hallucinatory responses. During inference, these steering vectors are added at all token positions after the user's prompt with either a positive or negative coefficient, allowing precise control over the degree of the targeted behavior. We evaluate CAA's effectiveness on Llama 2 Chat using multiple-choice behavioral question datasets and open-ended generation tasks. We demonstrate that CAA significantly alters model behavior, is effective over and on top of traditional methods like finetuning and system prompt design, and minimally reduces capabilities. Moreover, we gain deeper insights into CAA's mechanisms by employing various activation space interpretation methods. CAA accurately steers model outputs and sheds light on how high-level concepts are represented in Large Language Models (LLMs).
△ Less
Submitted 5 July, 2024; v1 submitted 8 December, 2023;
originally announced December 2023.
-
Understanding and Controlling a Maze-Solving Policy Network
Authors:
Ulisse Mini,
Peli Grietzer,
Mrinank Sharma,
Austin Meek,
Monte MacDiarmid,
Alexander Matt Turner
Abstract:
To understand the goals and goal representations of AI systems, we carefully study a pretrained reinforcement learning policy that solves mazes by navigating to a range of target squares. We find this network pursues multiple context-dependent goals, and we further identify circuits within the network that correspond to one of these goals. In particular, we identified eleven channels that track th…
▽ More
To understand the goals and goal representations of AI systems, we carefully study a pretrained reinforcement learning policy that solves mazes by navigating to a range of target squares. We find this network pursues multiple context-dependent goals, and we further identify circuits within the network that correspond to one of these goals. In particular, we identified eleven channels that track the location of the goal. By modifying these channels, either with hand-designed interventions or by combining forward passes, we can partially control the policy. We show that this network contains redundant, distributed, and retargetable goal representations, shedding light on the nature of goal-direction in trained policy networks.
△ Less
Submitted 12 October, 2023;
originally announced October 2023.
-
Optimally Localized Wannier Functions for 2D Chern Insulators
Authors:
Thivan M. Gunawardana,
Ari M. Turner,
Ryan Barnett
Abstract:
The construction of optimally localized Wannier functions (and Wannier functions in general) for a Chern insulator has been considered to be impossible owing to the fact that the second moment of such functions is generally infinite. In this manuscript, we propose a solution to this problem in the case of a single band. We accomplish this by drawing an analogy between the minimization of the varia…
▽ More
The construction of optimally localized Wannier functions (and Wannier functions in general) for a Chern insulator has been considered to be impossible owing to the fact that the second moment of such functions is generally infinite. In this manuscript, we propose a solution to this problem in the case of a single band. We accomplish this by drawing an analogy between the minimization of the variance and the minimization of the electrostatic energy of a periodic array of point charges in a smooth neutralizing background. In doing so, we obtain a natural regularization of the diverging variance and this leads to an analytical solution to the minimization problem. We demonstrate our results numerically for a particular model system. Furthermore, we show how the optimally localized Wannier functions provide a natural way of evaluating the electric polarization for a Chern insulator.
△ Less
Submitted 2 April, 2024; v1 submitted 13 September, 2023;
originally announced September 2023.
-
Activation Addition: Steering Language Models Without Optimization
Authors:
Alexander Matt Turner,
Lisa Thiergart,
Gavin Leech,
David Udell,
Juan J. Vazquez,
Ulisse Mini,
Monte MacDiarmid
Abstract:
Reliably controlling the behavior of large language models is a pressing open problem. Existing methods include supervised finetuning, reinforcement learning from human feedback, prompt engineering and guided decoding. We instead investigate activation engineering: modifying activations at inference-time to predictably alter model behavior. We bias the forward pass with a 'steering vector' implici…
▽ More
Reliably controlling the behavior of large language models is a pressing open problem. Existing methods include supervised finetuning, reinforcement learning from human feedback, prompt engineering and guided decoding. We instead investigate activation engineering: modifying activations at inference-time to predictably alter model behavior. We bias the forward pass with a 'steering vector' implicitly specified through natural language. Past work learned these steering vectors; our Activation Addition (ActAdd) method instead computes them by taking activation differences resulting from pairs of prompts. We demonstrate ActAdd on a range of LLMs (LLaMA-3, OPT, GPT-2, and GPT-J), obtaining SOTA on detoxification and negative-to-positive sentiment control. Our approach yields inference-time control over high-level properties of output like topic and sentiment while preserving performance on off-target tasks. ActAdd takes far less compute and implementation effort than finetuning or RLHF, allows users control through natural language, and its computational overhead (as a fraction of inference time) appears stable or improving over increasing model size.
△ Less
Submitted 4 June, 2024; v1 submitted 20 August, 2023;
originally announced August 2023.
-
Local and extensive fluctuations in sparsely-interacting ecological communities
Authors:
Stav Marcus,
Ari M Turner,
Guy Bunin
Abstract:
Ecological communities with many species can be classified into dynamical phases. In systems with all-to-all interactions, a phase where a fixed point is always reached and a dynamically-fluctuating phase have been found. The dynamics when interactions are sparse, with each species interacting with only several others, has remained largely unexplored. Here we show that a new type of phase appears…
▽ More
Ecological communities with many species can be classified into dynamical phases. In systems with all-to-all interactions, a phase where a fixed point is always reached and a dynamically-fluctuating phase have been found. The dynamics when interactions are sparse, with each species interacting with only several others, has remained largely unexplored. Here we show that a new type of phase appears in the phase diagram, where for the same control parameters different communities may reach either a fixed point or a state where the abundances of a finite subset of species fluctuate, and calculate the probability for each outcome. These fluctuating species are organized around short cycles in the interaction graph, and their abundances undergo large non-linear fluctuations. We characterize the approach from this phase to a phase with extensively many fluctuating species, and show that the probability of fluctuations grows continuously to one as the transition is approached, and that the number of fluctuating species diverges. This is qualitatively distinct from the transition to extensive fluctuations coming from a fixed point phase, which is marked by a loss of linear stability. The differences are traced back to the emergent binary character of the dynamics when far away from short cycles in the local fluctuations phase.
△ Less
Submitted 3 August, 2023;
originally announced August 2023.
-
Homotopy of periodic two by two matrices
Authors:
Joseph E. Avron,
Ari M. Turner
Abstract:
We describe the homotopy classes of 2 by 2 periodic simple (=non-degenerate) matrices with various symmetries. This turns out to be an elementary exercise in the homotopy of closed curves in three dimensions. The matrices represent gapped Bloch Hamiltonians in 1D with a two-dimensional Hilbert space per unit cell.
We describe the homotopy classes of 2 by 2 periodic simple (=non-degenerate) matrices with various symmetries. This turns out to be an elementary exercise in the homotopy of closed curves in three dimensions. The matrices represent gapped Bloch Hamiltonians in 1D with a two-dimensional Hilbert space per unit cell.
△ Less
Submitted 20 February, 2024; v1 submitted 14 December, 2022;
originally announced December 2022.
-
Generalized Lotka-Volterra equations with random, non-reciprocal interactions: the typical number of equilibria
Authors:
Valentina Ros,
Felix Roy,
Giulio Biroli,
Guy Bunin,
Ari M. Turner
Abstract:
We compute the typical number of equilibria of the Generalized Lotka-Volterra equations describing species-rich ecosystems with random, non-reciprocal interactions using the replicated Kac-Rice method. We characterize the multiple-equilibria phase by determining the average abundance and similaritybetween equilibria as a function of their diversity (i.e. of the number of coexisting species) and of…
▽ More
We compute the typical number of equilibria of the Generalized Lotka-Volterra equations describing species-rich ecosystems with random, non-reciprocal interactions using the replicated Kac-Rice method. We characterize the multiple-equilibria phase by determining the average abundance and similaritybetween equilibria as a function of their diversity (i.e. of the number of coexisting species) and of the variability of the interactions. We show that linearly unstable equilibria are dominant, and that the typical number of equilibria differs with respect to the average number.
△ Less
Submitted 25 June, 2023; v1 submitted 4 December, 2022;
originally announced December 2022.
-
Parametrically Retargetable Decision-Makers Tend To Seek Power
Authors:
Alexander Matt Turner,
Prasad Tadepalli
Abstract:
If capable AI agents are generally incentivized to seek power in service of the objectives we specify for them, then these systems will pose enormous risks, in addition to enormous benefits. In fully observable environments, most reward functions have an optimal policy which seeks power by kee** options open and staying alive. However, the real world is neither fully observable, nor must trained…
▽ More
If capable AI agents are generally incentivized to seek power in service of the objectives we specify for them, then these systems will pose enormous risks, in addition to enormous benefits. In fully observable environments, most reward functions have an optimal policy which seeks power by kee** options open and staying alive. However, the real world is neither fully observable, nor must trained agents be even approximately reward-optimal. We consider a range of models of AI decision-making, from optimal, to random, to choices informed by learning and interacting with an environment. We discover that many decision-making functions are retargetable, and that retargetability is sufficient to cause power-seeking tendencies. Our functional criterion is simple and broad. We show that a range of qualitatively dissimilar decision-making procedures incentivize agents to seek power. We demonstrate the flexibility of our results by reasoning about learned policy incentives in Montezuma's Revenge. These results suggest a safety risk: Eventually, retargetable training procedures may train real-world agents which seek power over humans.
△ Less
Submitted 11 October, 2022; v1 submitted 27 June, 2022;
originally announced June 2022.
-
On Avoiding Power-Seeking by Artificial Intelligence
Authors:
Alexander Matt Turner
Abstract:
We do not know how to align a very intelligent AI agent's behavior with human interests. I investigate whether -- absent a full solution to this AI alignment problem -- we can build smart AI agents which have limited impact on the world, and which do not autonomously seek power. In this thesis, I introduce the attainable utility preservation (AUP) method. I demonstrate that AUP produces conservati…
▽ More
We do not know how to align a very intelligent AI agent's behavior with human interests. I investigate whether -- absent a full solution to this AI alignment problem -- we can build smart AI agents which have limited impact on the world, and which do not autonomously seek power. In this thesis, I introduce the attainable utility preservation (AUP) method. I demonstrate that AUP produces conservative, option-preserving behavior within toy gridworlds and within complex environments based off of Conway's Game of Life. I formalize the problem of side effect avoidance, which provides a way to quantify the side effects an agent had on the world. I also give a formal definition of power-seeking in the context of AI agents and show that optimal policies tend to seek power. In particular, most reward functions have optimal policies which avoid deactivation. This is a problem if we want to deactivate or correct an intelligent agent after we have deployed it. My theorems suggest that since most agent goals conflict with ours, the agent would very probably resist correction. I extend these theorems to show that power-seeking incentives occur not just for optimal decision-makers, but under a wide range of decision-making procedures.
△ Less
Submitted 23 June, 2022;
originally announced June 2022.
-
Formalizing the Problem of Side Effect Regularization
Authors:
Alexander Matt Turner,
Aseem Saxena,
Prasad Tadepalli
Abstract:
AI objectives are often hard to specify properly. Some approaches tackle this problem by regularizing the AI's side effects: Agents must weigh off "how much of a mess they make" with an imperfectly specified proxy objective. We propose a formal criterion for side effect regularization via the assistance game framework. In these games, the agent solves a partially observable Markov decision process…
▽ More
AI objectives are often hard to specify properly. Some approaches tackle this problem by regularizing the AI's side effects: Agents must weigh off "how much of a mess they make" with an imperfectly specified proxy objective. We propose a formal criterion for side effect regularization via the assistance game framework. In these games, the agent solves a partially observable Markov decision process (POMDP) representing its uncertainty about the objective function it should optimize. We consider the setting where the true objective is revealed to the agent at a later time step. We show that this POMDP is solved by trading off the proxy reward with the agent's ability to achieve a range of future tasks. We empirically demonstrate the reasonableness of our problem formalization via ground-truth evaluation in two gridworld environments.
△ Less
Submitted 8 November, 2022; v1 submitted 23 June, 2022;
originally announced June 2022.
-
Long-range influence of a pump on a critical fluid
Authors:
Ydan Ben Dor,
Yariv Kafri,
David Mukamel,
Ari M. Turner
Abstract:
A pump coupled to a conserved density generates long-range modulations, resulting from the non-equilibrium nature of the dynamics. We study how these modulations are modified at the critical point where the system exhibits intrinsic long-range correlations. To do so, we consider a pump in a diffusive fluid, which is known to generate a density profile in the form of an electric dipole potential an…
▽ More
A pump coupled to a conserved density generates long-range modulations, resulting from the non-equilibrium nature of the dynamics. We study how these modulations are modified at the critical point where the system exhibits intrinsic long-range correlations. To do so, we consider a pump in a diffusive fluid, which is known to generate a density profile in the form of an electric dipole potential and a current in the form of a dipolar field above the critical point. We demonstrate that while the current retains its form at the critical point, the density profile changes drastically. At criticality, in $d<4$ dimensions, the deviation of the density from the average is given by ${\rm sgn}(\cos(θ))|\cos(θ)/r^{(d-1)}|^{1/δ}$ at large distance $r$ from the pump and angle $θ$ with respect to the pump's orientation. At short distances, there is a crossover to a $\cos(θ)/r^{d-3+η}$ profile. Here $δ$ and $η$ are Ising critical exponents. The effect of the local pump on the domain wall structure below the critical point is also considered.
△ Less
Submitted 29 March, 2022; v1 submitted 29 November, 2021;
originally announced November 2021.
-
Local and collective transitions in sparsely-interacting ecological communities
Authors:
Stav Marcus,
Ari M. Turner,
Guy Bunin
Abstract:
Interactions in natural communities can be highly heterogeneous, with any given species interacting appreciably with only some of the others, a situation commonly represented by sparse interaction networks. We study the consequences of sparse competitive interactions, in a theoretical model of a community assembled from a species pool. We find that communities can be in a number of different regim…
▽ More
Interactions in natural communities can be highly heterogeneous, with any given species interacting appreciably with only some of the others, a situation commonly represented by sparse interaction networks. We study the consequences of sparse competitive interactions, in a theoretical model of a community assembled from a species pool. We find that communities can be in a number of different regimes, depending on the interaction strength. When interactions are strong, the network of coexisting species breaks up into small subgraphs, while for weaker interactions these graphs are larger and more complex, eventually encompassing all species. This process is driven by emergence of new allowed subgraphs as interaction strength decreases, leading to sharp changes in diversity and other community properties, and at weaker interactions to two distinct collective transitions: a percolation transition, and a transition between having a unique equilibrium and having multiple alternative equilibria. Understanding community structure is thus made up of two parts: first, finding which subgraphs are allowed at a given interaction strength, and secondly, a discrete problem of matching these structures over the entire community. In a shift from the focus of many previous theories, these different regimes can be traversed by modifying the interaction strength alone, without need for heterogeneity in either interaction strengths or the number of competitors per species.
△ Less
Submitted 26 October, 2021;
originally announced October 2021.
-
Gap** fragile topological bands by interactions
Authors:
Ari M. Turner,
Erez Berg,
Ady Stern
Abstract:
We consider the stability of fragile topological bands protected by space-time inversion symmetry in the presence of strong electron-electron interactions. At the single-particle level, the topological nature of the bands prevents the opening of a gap between them. In contrast, we show that when the fragile bands are half filled, interactions can open a gap in the many-body spectrum without breaki…
▽ More
We consider the stability of fragile topological bands protected by space-time inversion symmetry in the presence of strong electron-electron interactions. At the single-particle level, the topological nature of the bands prevents the opening of a gap between them. In contrast, we show that when the fragile bands are half filled, interactions can open a gap in the many-body spectrum without breaking any symmetry or mixing degrees of freedom from remote bands. Furthermore, the resulting ground state is not topologically ordered. Thus, a fragile topological band structure does not present an obstruction to forming a "featureless insulator" ground state. Our construction relies on the formation of fermionic bound states of two electrons and one hole, known as "trions". The trions form a band whose coupling to the electronic band enables the gap opening. This result may be relevant to the gapped state indicated by recent experiments in magic angle twisted bilayer graphene at charge neutrality.
△ Less
Submitted 18 February, 2022; v1 submitted 19 April, 2021;
originally announced April 2021.
-
Origin and localization of topological band gaps in gyroscopic metamaterials
Authors:
Noah P. Mitchell,
Ari M. Turner,
William T. M. Irvine
Abstract:
Networks of interacting gyroscopes have proven to be versatile structures for understanding and harnessing finite-frequency topological excitations. Spinning components give rise to band gaps and topologically protected wave transport along the system's boundaries, whether the gyroscopes are arranged in a lattice or in an amorphous configuration. Here, we examine the irrelevance of periodic order…
▽ More
Networks of interacting gyroscopes have proven to be versatile structures for understanding and harnessing finite-frequency topological excitations. Spinning components give rise to band gaps and topologically protected wave transport along the system's boundaries, whether the gyroscopes are arranged in a lattice or in an amorphous configuration. Here, we examine the irrelevance of periodic order for generating topological gaps. Starting from the symplectic dynamics of our model metamaterial, we present a general method for predicting whether a gap exists and for approximating the Chern number using only local features of a network, bypassing the costly diagonalization of the system's dynamical matrix. We then study how strong disorder interacts with band topology in gyroscopic metamaterials and find that amorphous gyroscopic Chern insulators exhibit similar critical behavior to periodic lattices. Our experiments and simulations additionally reveal a topological Anderson insulation transition, wherein disorder drives a trivial phase into a topological one.
△ Less
Submitted 16 December, 2020;
originally announced December 2020.
-
Avoiding Side Effects in Complex Environments
Authors:
Alexander Matt Turner,
Neale Ratzlaff,
Prasad Tadepalli
Abstract:
Reward function specification can be difficult. Rewarding the agent for making a widget may be easy, but penalizing the multitude of possible negative side effects is hard. In toy environments, Attainable Utility Preservation (AUP) avoided side effects by penalizing shifts in the ability to achieve randomly generated goals. We scale this approach to large, randomly generated environments based on…
▽ More
Reward function specification can be difficult. Rewarding the agent for making a widget may be easy, but penalizing the multitude of possible negative side effects is hard. In toy environments, Attainable Utility Preservation (AUP) avoided side effects by penalizing shifts in the ability to achieve randomly generated goals. We scale this approach to large, randomly generated environments based on Conway's Game of Life. By preserving optimal value for a single randomly generated reward function, AUP incurs modest overhead while leading the agent to complete the specified task and avoid many side effects. Videos and code are available at https://avoiding-side-effects.github.io/.
△ Less
Submitted 22 October, 2020; v1 submitted 11 June, 2020;
originally announced June 2020.
-
Neutron scattering off Weyl semimetals
Authors:
Michael Bjerngaard,
Bogdan Galilo,
Ari M. Turner
Abstract:
We present how to detect type-$1$ Weyl nodes in a material by inelastic neutron scattering. Such an experiment first of all allows one to determine the dispersion of the Weyl fermions. We extend the reasoning to produce a quantitative test of the Weyl equation taking into account realistic anisotropic properties. These anisotropies are mostly contained in the form of the emergent magnetic moment o…
▽ More
We present how to detect type-$1$ Weyl nodes in a material by inelastic neutron scattering. Such an experiment first of all allows one to determine the dispersion of the Weyl fermions. We extend the reasoning to produce a quantitative test of the Weyl equation taking into account realistic anisotropic properties. These anisotropies are mostly contained in the form of the emergent magnetic moment of the excitations, which determines how they couple to the neutron. Although there are many material parameters, we find several quantitative predictions that are universal and demonstrate that the excitations are described by solutions to the Weyl equation. The realistic, anisotropic coupling between electrons and neutrons implies that even fully unpolarized neutrons can reveal the spin-momentum locking of the Weyl fermions because the neutrons will couple to some components of the Weyl fermion pseudospin more strongly. On the other hand, in an experiment with polarized neutrons, the scattered neutron beam remains fully polarized in a direction that varies as a function of momentum transfer (within the range of validity of the Weyl equation). This allows measurement of the chirality of Weyl fermions for inversion symmetric nodes. Furthermore, we estimate that the scattering rate may be large enough for such experiments to be practical; in particular, the magnetic moment may be larger than the ordinary Bohr magneton, compensating for a small density of states.
△ Less
Submitted 13 March, 2020; v1 submitted 16 February, 2020;
originally announced February 2020.
-
Optimal Policies Tend to Seek Power
Authors:
Alexander Matt Turner,
Logan Smith,
Rohin Shah,
Andrew Critch,
Prasad Tadepalli
Abstract:
Some researchers speculate that intelligent reinforcement learning (RL) agents would be incentivized to seek resources and power in pursuit of their objectives. Other researchers point out that RL agents need not have human-like power-seeking instincts. To clarify this discussion, we develop the first formal theory of the statistical tendencies of optimal policies. In the context of Markov decisio…
▽ More
Some researchers speculate that intelligent reinforcement learning (RL) agents would be incentivized to seek resources and power in pursuit of their objectives. Other researchers point out that RL agents need not have human-like power-seeking instincts. To clarify this discussion, we develop the first formal theory of the statistical tendencies of optimal policies. In the context of Markov decision processes, we prove that certain environmental symmetries are sufficient for optimal policies to tend to seek power over the environment. These symmetries exist in many environments in which the agent can be shut down or destroyed. We prove that in these environments, most reward functions make it optimal to seek power by kee** a range of options available and, when maximizing average reward, by navigating towards larger sets of potential terminal states.
△ Less
Submitted 28 January, 2023; v1 submitted 3 December, 2019;
originally announced December 2019.
-
Conservative Agency via Attainable Utility Preservation
Authors:
Alexander Matt Turner,
Dylan Hadfield-Menell,
Prasad Tadepalli
Abstract:
Reward functions are easy to misspecify; although designers can make corrections after observing mistakes, an agent pursuing a misspecified reward function can irreversibly change the state of its environment. If that change precludes optimization of the correctly specified reward function, then correction is futile. For example, a robotic factory assistant could break expensive equipment due to a…
▽ More
Reward functions are easy to misspecify; although designers can make corrections after observing mistakes, an agent pursuing a misspecified reward function can irreversibly change the state of its environment. If that change precludes optimization of the correctly specified reward function, then correction is futile. For example, a robotic factory assistant could break expensive equipment due to a reward misspecification; even if the designers immediately correct the reward function, the damage is done. To mitigate this risk, we introduce an approach that balances optimization of the primary reward function with preservation of the ability to optimize auxiliary reward functions. Surprisingly, even when the auxiliary reward functions are randomly generated and therefore uninformative about the correctly specified reward function, this approach induces conservative, effective behavior.
△ Less
Submitted 10 June, 2020; v1 submitted 25 February, 2019;
originally announced February 2019.
-
Statistics of fractionalized excitations through threshold spectroscopy
Authors:
Siddhardh C. Morampudi,
Ari M. Turner,
Frank Pollmann,
Frank Wilczek
Abstract:
We show that the anyonic statistics of fractionalized excitations display characteristic signatures in threshold spectroscopic measurements. Drawing motivation from topologically ordered phases such as gapped quantum spin liquids and fractional chern insulators which possess fractionalized excitations, we consider gapped systems with abelian anyonic excitations. The low energy onset of associated…
▽ More
We show that the anyonic statistics of fractionalized excitations display characteristic signatures in threshold spectroscopic measurements. Drawing motivation from topologically ordered phases such as gapped quantum spin liquids and fractional chern insulators which possess fractionalized excitations, we consider gapped systems with abelian anyonic excitations. The low energy onset of associated correlation functions near the threshold show universal behaviour depending on the statistics of the anyons. This explains some recent theoretical results in spin systems and also provides a route towards detecting statistics in experiments such as neutron scattering and tunneling spectroscopy.
△ Less
Submitted 19 August, 2016;
originally announced August 2016.
-
Topological mechanics of gyroscopic metamaterials
Authors:
Lisa M. Nash,
Dustin Kleckner,
Alismari Read,
Vincenzo Vitelli,
Ari M. Turner,
William T. M. Irvine
Abstract:
Topological mechanical metamaterials are artificial structures whose unusual properties are protected very much like their electronic and optical counterparts. Here, we present an experimental and theoretical study of an active metamaterial -- comprised of coupled gyroscopes on a lattice -- that breaks time-reversal symmetry. The vibrational spectrum of these novel structures displays a sonic gap…
▽ More
Topological mechanical metamaterials are artificial structures whose unusual properties are protected very much like their electronic and optical counterparts. Here, we present an experimental and theoretical study of an active metamaterial -- comprised of coupled gyroscopes on a lattice -- that breaks time-reversal symmetry. The vibrational spectrum of these novel structures displays a sonic gap populated by topologically protected edge modes which propagate in only one direction and are unaffected by disorder. We present a mathematical model that explains how the edge mode chirality can be switched via controlled distortions of the underlying lattice. This effect allows the direction of the edge current to be determined on demand. We envision applications of these edges modes to the design of loss-free, one-way, acoustic waveguides and demonstrate this functionality in experiment.
△ Less
Submitted 17 April, 2015; v1 submitted 13 April, 2015;
originally announced April 2015.
-
Geometrical detection of weak non-Gaussianity upon coarse-graining
Authors:
T. H. Beuman,
A. M. Turner,
V. Vitelli
Abstract:
Measures of the non-Gaussianity of a random field depend on how accurately one is able to measure the field. If a signal measured at a certain point is to be averaged with its surroundings, or coarse-grained, the magnitude of its non-Gaussian component can vary. In this article, we investigate the variation of the "apparent" non-Gaussianity, as a function of the coarse-graining length, when we mea…
▽ More
Measures of the non-Gaussianity of a random field depend on how accurately one is able to measure the field. If a signal measured at a certain point is to be averaged with its surroundings, or coarse-grained, the magnitude of its non-Gaussian component can vary. In this article, we investigate the variation of the "apparent" non-Gaussianity, as a function of the coarse-graining length, when we measure non-Gaussianity using the statistics of extrema in the field. We derive how the relative difference between maxima and minima -- which is a geometrical measure of the field's non-Gaussianity -- behaves as the field is coarse-grained over increasingly larger length scales. Measuring this function can give extra information about the non-Gaussian statistics and facilitate its detection.
△ Less
Submitted 27 February, 2014;
originally announced February 2014.
-
Berry phase induced dimerization in one-dimensional quadrupolar systems
Authors:
Shijie Hu,
Ari M. Turner,
Karlo Penc,
Frank Pollmann
Abstract:
We investigate the effect of the Berry phase on quadrupoles that occur for example in the low-energy description of spin models. Specifically we study here the one-dimensional bilinear-biquadratic spin-one model. An open question for many years about this model is whether it has a non-dimerized fluctuating nematic phase. The dimerization has recently been proposed to be related to Berry phases of…
▽ More
We investigate the effect of the Berry phase on quadrupoles that occur for example in the low-energy description of spin models. Specifically we study here the one-dimensional bilinear-biquadratic spin-one model. An open question for many years about this model is whether it has a non-dimerized fluctuating nematic phase. The dimerization has recently been proposed to be related to Berry phases of the quantum fluctuations. We use an effective low-energy description to calculate the scaling of the dimerization according to this theory, and then verify the predictions using large scale density-matrix renormalization group (DMRG) simulations, giving good evidence that the state is dimerized all the way up to its transition into the ferromagnetic phase. We furthermore discuss the multiplet structure found in the entanglement spectrum of the ground state wave functions.
△ Less
Submitted 10 June, 2014; v1 submitted 14 January, 2014;
originally announced January 2014.
-
Solitons and thermal fluctuations in strongly nonlinear solids
Authors:
N. Upadhyaya,
A. M. Turner,
V. Vitelli
Abstract:
We study a chain of anharmonic springs with tunable power law interactions as a minimal model to explore the propagation of strongly non-linear solitary wave excitations in a background of thermal fluctuations. By treating the solitary waves as quasi-particles, we derive an effective Langevin equation and obtain their dam** rate and thermal diffusion. These analytical findings compare favorably…
▽ More
We study a chain of anharmonic springs with tunable power law interactions as a minimal model to explore the propagation of strongly non-linear solitary wave excitations in a background of thermal fluctuations. By treating the solitary waves as quasi-particles, we derive an effective Langevin equation and obtain their dam** rate and thermal diffusion. These analytical findings compare favorably against numerical results from a Langevin dynamic simulation. In our chains composed of two sided non-linear springs, we report the existence of an expansion solitary wave (anti-soliton) in addition to the compressive solitary waves observed for non-cohesive macroscopic particles.
△ Less
Submitted 11 October, 2013; v1 submitted 24 April, 2013;
originally announced April 2013.
-
Topological phases in gapped edges of fractionalized systems
Authors:
Johannes Motruk,
Ari M. Turner,
Erez Berg,
Frank Pollmann
Abstract:
Recently, it has been proposed that exotic one-dimensional phases can be realized by gap** out the edge states of a fractional topological insulator. The low-energy edge degrees of freedom are described by a chain of coupled parafermions. We introduce a classification scheme for the phases that can occur in parafermionic chains. We find that the parafermions support both topological symmetry fra…
▽ More
Recently, it has been proposed that exotic one-dimensional phases can be realized by gap** out the edge states of a fractional topological insulator. The low-energy edge degrees of freedom are described by a chain of coupled parafermions. We introduce a classification scheme for the phases that can occur in parafermionic chains. We find that the parafermions support both topological symmetry fractionalized phases as well as phases in which the parafermions condense. In the presence of additional symmetries, the phases form a non-Abelian group. As a concrete example of the classification, we consider the effective edge model for a $ν= 1/3$ fractional topological insulator for which we calculate the entanglement spectra numerically and show that all possible predicted phases can be realized.
△ Less
Submitted 10 October, 2013; v1 submitted 9 March, 2013;
originally announced March 2013.
-
Beyond Band Insulators: Topology of Semi-metals and Interacting Phases
Authors:
Ari M. Turner,
Ashvin Vishwanath
Abstract:
The theory of topological insulators and superconductors has mostly focused on non-interacting and gapped systems. This review article discusses topological phases that are either gapless or interacting. We discuss recent progress in identifying gapless systems with stable topological properties (such as novel surface states), using Weyl semimetals as an illustration. We then review recent progres…
▽ More
The theory of topological insulators and superconductors has mostly focused on non-interacting and gapped systems. This review article discusses topological phases that are either gapless or interacting. We discuss recent progress in identifying gapless systems with stable topological properties (such as novel surface states), using Weyl semimetals as an illustration. We then review recent progress in describing topological phases of interacting gapped systems. We explain how new types of edge states can be stabilized by interactions and symmetry, even though the bulk has only conventional excitations and no topological order of the kind associated with Fractional Quantum Hall states.
△ Less
Submitted 2 January, 2013;
originally announced January 2013.
-
Topological Order and Absence of Band Insulators at Integer Filling in Non-Symmorphic Crystals
Authors:
S. A. Parameswaran,
Ari M. Turner,
Daniel P. Arovas,
Ashvin Vishwanath
Abstract:
Band insulators appear in a crystalline system only when the filling -- the number of electrons per unit cell and spin projection -- is an integer. At fractional filling, an insulating phase that preserves all symmetries is a Mott insulator, i.e. it is either gapless or, if gapped, displays fractionalized excitations and topological order. We raise the inverse question -- at an integer filling is…
▽ More
Band insulators appear in a crystalline system only when the filling -- the number of electrons per unit cell and spin projection -- is an integer. At fractional filling, an insulating phase that preserves all symmetries is a Mott insulator, i.e. it is either gapless or, if gapped, displays fractionalized excitations and topological order. We raise the inverse question -- at an integer filling is a band insulator always possible? Here we show that lattice symmetries may forbid a band insulator even at certain integer fillings, if the crystal is non-symmorphic -- a property shared by a majority of three-dimensional crystal structures. In these cases, one may infer the existence of topological order if the ground state is gapped and fully symmetric. This is demonstrated using a non-perturbative flux threading argument, which has immediate applications to quantum spin systems and bosonic insulators in addition to electronic band structures in the absence of spin-orbit interactions.
△ Less
Submitted 10 December, 2012; v1 submitted 3 December, 2012;
originally announced December 2012.
-
Extrema statistics in the dynamics of a non-Gaussian random field
Authors:
T. H. Beuman,
A. M. Turner,
V. Vitelli
Abstract:
When the equations that govern the dynamics of a random field are nonlinear, the field can develop with time non-Gaussian statistics even if its initial condition is Gaussian. Here, we provide a general framework for calculating the effect of the underlying nonlinear dynamics on the relative densities of maxima and minima of the field. Using this simple geometrical probe, we can identify the size…
▽ More
When the equations that govern the dynamics of a random field are nonlinear, the field can develop with time non-Gaussian statistics even if its initial condition is Gaussian. Here, we provide a general framework for calculating the effect of the underlying nonlinear dynamics on the relative densities of maxima and minima of the field. Using this simple geometrical probe, we can identify the size of the non-Gaussian contributions in the random field, or alternatively the magnitude of the nonlinear terms in the underlying equations of motion. We demonstrate our approach by applying it to an initially Gaussian field that evolves according to the deterministic KPZ equation, which models surface growth and shock dynamics.
△ Less
Submitted 5 November, 2012;
originally announced November 2012.
-
Umbilical points of a non-Gaussian random field
Authors:
A. M. Turner,
T. H. Beuman,
V. Vitelli
Abstract:
Random fields in nature often have, to a good approximation, Gaussian characteristics. For such fields, the relative densities of umbilical points -- topological defects which can be classified into three types -- have certain fixed values. Phenomena described by nonlinear laws can however give rise to a non-Gaussian contribution, causing a deviation from these universal values. We consider a Gaus…
▽ More
Random fields in nature often have, to a good approximation, Gaussian characteristics. For such fields, the relative densities of umbilical points -- topological defects which can be classified into three types -- have certain fixed values. Phenomena described by nonlinear laws can however give rise to a non-Gaussian contribution, causing a deviation from these universal values. We consider a Gaussian field with a perturbation added to it, given by a nonlinear function of that field, and calculate the change in the relative density of umbilical points. This allows us not only to detect a perturbation, but to determine its size as well. This geometric approach offers an independent way of detecting non-Gaussianity, which even works in cases where the field itself cannot be probed directly.
△ Less
Submitted 3 November, 2012;
originally announced November 2012.
-
Critical points of a non-Gaussian random field
Authors:
T. H. Beuman,
A. M. Turner,
V. Vitelli
Abstract:
Random fields in nature often have, to a good approximation, Gaussian characteristics. We present the mathematical framework for a new and simple method for investigating the non-Gaussian contributions, based on counting the maxima and minima of a scalar field. We consider a random surface, whose height is given by a nonlinear function of a Gaussian field. We find that, as a result of the non-Gaus…
▽ More
Random fields in nature often have, to a good approximation, Gaussian characteristics. We present the mathematical framework for a new and simple method for investigating the non-Gaussian contributions, based on counting the maxima and minima of a scalar field. We consider a random surface, whose height is given by a nonlinear function of a Gaussian field. We find that, as a result of the non-Gaussianity, the density of maxima and minima no longer match and calculate the relative imbalance between the two. Our approach allows to detect and quantify non-Gaussianities present in any random field that can be represented as the height of a smooth two-dimensional surface.
△ Less
Submitted 25 October, 2012;
originally announced October 2012.
-
Uniform Shock Waves in Disordered Granular Matter
Authors:
L. R. Gomez,
A. M. Turner,
V. Vitelli
Abstract:
The confining pressure $P$ is perhaps the most important parameter controlling the properties of granular matter. Strongly compressed granular media are, in many respects, simple solids in which elastic perturbations travel as ordinary phonons. However, the speed of sound in granular aggregates continuously decreases as the confining pressure decreases, completely vanishing at the jamming-unjammin…
▽ More
The confining pressure $P$ is perhaps the most important parameter controlling the properties of granular matter. Strongly compressed granular media are, in many respects, simple solids in which elastic perturbations travel as ordinary phonons. However, the speed of sound in granular aggregates continuously decreases as the confining pressure decreases, completely vanishing at the jamming-unjamming transition. This anomalous behavior suggests that the transport of energy at low pressures should not be dominated by phonons. In this work we use simulations and theory to show how the response of granular systems becomes increasingly nonlinear as pressure decreases. In the low pressure regime the elastic energy is found to be mainly transported through nonlinear waves and shocks. We numerically characterize the propagation speed, shape, and stability of these shocks, and model the dependence of the shock speed on pressure and impact intensity by a simple analytical approach.
△ Less
Submitted 1 August, 2012;
originally announced August 2012.
-
Stochastic geometry and topology of non-Gaussian fields
Authors:
T. H. Beuman,
A. M. Turner,
V. Vitelli
Abstract:
Gaussian random fields pervade all areas of science. However, it is often the departures from Gaussianity that carry the crucial signature of the nonlinear mechanisms at the heart of diverse phenomena, ranging from structure formation in condensed matter and cosmology to biomedical imaging. The standard test of non-Gaussianity is to measure higher order correlation functions. In the present work,…
▽ More
Gaussian random fields pervade all areas of science. However, it is often the departures from Gaussianity that carry the crucial signature of the nonlinear mechanisms at the heart of diverse phenomena, ranging from structure formation in condensed matter and cosmology to biomedical imaging. The standard test of non-Gaussianity is to measure higher order correlation functions. In the present work, we take a different route. We show how geometric and topological properties of Gaussian fields, such as the statistics of extrema, are modified by the presence of a non-Gaussian perturbation. The resulting discrepancies give an independent way to detect and quantify non-Gaussianities. In our treatment, we consider both local and nonlocal mechanisms that generate non-Gaussian fields, both statically and dynamically through nonlinear diffusion.
△ Less
Submitted 17 July, 2012;
originally announced July 2012.
-
Featureless and non-fractionalized Mott insulators on the honeycomb lattice at 1/2 site filling
Authors:
Itamar Kimchi,
S. A. Parameswaran,
Ari M. Turner,
Fa Wang,
Ashvin Vishwanath
Abstract:
Within the Landau paradigm, phases of matter are distinguished by spontaneous symmetry breaking. Implicit here is the assumption that a completely symmetric state exists: a paramagnet. At zero temperature such quantum featureless insulators may be forbidden, triggering either conventional order or topological order with fractionalized excitations. Such is the case for interacting particles when th…
▽ More
Within the Landau paradigm, phases of matter are distinguished by spontaneous symmetry breaking. Implicit here is the assumption that a completely symmetric state exists: a paramagnet. At zero temperature such quantum featureless insulators may be forbidden, triggering either conventional order or topological order with fractionalized excitations. Such is the case for interacting particles when the particle number per unit cell, f, is not an integer. But, can lattice symmetries forbid featureless insulators even at integer f? An especially relevant case is the honeycomb (graphene) lattice --- where free spinless fermions at f=1 (the two sites per unit cell mean f=1 is half filling per site) are always metallic. Here we present wave functions for bosons, and a related spin-singlet wave function for spinful electrons, on the f=1 honeycomb, and demonstrate via quantum to classical map**s that they do form featureless Mott insulators. The construction generalizes to symmorphic lattices at integer f in any dimension. Our results explicitly demonstrate that in this case, despite the absence of a non-interacting insulator at the same filling, lack of order at zero temperature does not imply fractionalization.
△ Less
Submitted 5 November, 2013; v1 submitted 2 July, 2012;
originally announced July 2012.
-
Wannier Permanent Wave Functions and Featureless Bosonic Mott Insulators on the 1/3 filled Kagome Lattice
Authors:
S. A. Parameswaran,
Itamar Kimchi,
Ari M. Turner,
D. M. Stamper-Kurn,
Ashvin Vishwanath
Abstract:
We study Bose-Hubbard models on tight-binding, non-Bravais lattices, with a filling of one boson per unit cell -- and thus fractional site filling. At integer filling of a unit cell neither symmetry breaking nor topological order is required, and in principle a trivial and featureless (i.e., symmetry-unbroken) insulator is allowed. We demonstrate by explicit construction of a family of wavefunctio…
▽ More
We study Bose-Hubbard models on tight-binding, non-Bravais lattices, with a filling of one boson per unit cell -- and thus fractional site filling. At integer filling of a unit cell neither symmetry breaking nor topological order is required, and in principle a trivial and featureless (i.e., symmetry-unbroken) insulator is allowed. We demonstrate by explicit construction of a family of wavefunctions that such a featureless Mott insulating state exists at 1/3 filling on the kagome lattice, and construct Hamiltonians for which these wavefunctions are exact ground states. We briefly comment on the experimental relevance of our results to cold atoms in optical lattices. Such wavefunctions also yield 1/3 magnetization plateau states for spin models in an applied field. The featureless Mott states we discuss can be generalized to any lattice for which symmetric exponentially localized Wannier orbitals can be found at the requisite filling, and their wavefunction is given by the permanent over all Wannier orbitals.
△ Less
Submitted 21 March, 2013; v1 submitted 5 June, 2012;
originally announced June 2012.
-
Detection of Symmetry Protected Topological Phases in 1D
Authors:
Frank Pollmann,
Ari M. Turner
Abstract:
A topological phase is a phase of matter which cannot be characterized by a local order parameter. It has been shown that gapped phases in 1D systems can be completely characterized using tools related to projective representations of the symmetry groups. We show how to determine the matrices of these representations in a simple way in order to distinguish between different phases directly. From t…
▽ More
A topological phase is a phase of matter which cannot be characterized by a local order parameter. It has been shown that gapped phases in 1D systems can be completely characterized using tools related to projective representations of the symmetry groups. We show how to determine the matrices of these representations in a simple way in order to distinguish between different phases directly. From these matrices we also point out how to derive several different types of non-local order parameters for time reversal, inversion symmetry and $Z_2 \times Z_2$ symmetry, as well as some more general cases (some of which have been obtained before by other methods). Using these concepts, the ordinary string order for the Haldane phase can be related to a selection rule that changes at the critical point. We furthermore point out an example of a more complicated internal symmetry for which the ordinary string order cannot be applied.
△ Less
Submitted 3 April, 2012;
originally announced April 2012.
-
Shocks near Jamming
Authors:
Leopoldo R. Gomez,
Ari M. Turner,
Martin van Hecke,
Vincenzo Vitelli
Abstract:
Non-linear sound is an extreme phenomenon typically observed in solids after violent explosions. But granular media are different. Right when they jam, these fragile and disordered solids exhibit a vanishing rigidity and sound speed, so that even tiny mechanical perturbations form supersonic shocks. Here, we perform simulations in which two-dimensional jammed granular packings are dynamically comp…
▽ More
Non-linear sound is an extreme phenomenon typically observed in solids after violent explosions. But granular media are different. Right when they jam, these fragile and disordered solids exhibit a vanishing rigidity and sound speed, so that even tiny mechanical perturbations form supersonic shocks. Here, we perform simulations in which two-dimensional jammed granular packings are dynamically compressed, and demonstrate that the elementary excitations are strongly non-linear shocks, rather than ordinary phonons. We capture the full dependence of the shock speed on pressure and impact intensity by a surprisingly simple analytical model.
△ Less
Submitted 7 December, 2011; v1 submitted 29 August, 2011;
originally announced August 2011.
-
Entanglement Entropy of Gapped Phases and Topological Order in Three dimensions
Authors:
Tarun Grover,
Ari M. Turner,
Ashvin Vishwanath
Abstract:
We discuss entanglement entropy of gapped ground states in different dimensions, obtained on partitioning space into two regions. For trivial phases without topological order, we argue that the entanglement entropy may be obtained by integrating an `entropy density' over the partition boundary that admits a gradient expansion in the curvature of the boundary. This constrains the expansion of entan…
▽ More
We discuss entanglement entropy of gapped ground states in different dimensions, obtained on partitioning space into two regions. For trivial phases without topological order, we argue that the entanglement entropy may be obtained by integrating an `entropy density' over the partition boundary that admits a gradient expansion in the curvature of the boundary. This constrains the expansion of entanglement entropy as a function of system size, and points to an even-odd dependence on dimensionality. For example, in contrast to the familiar result in two dimensions, a size independent constant contribution to the entanglement entropy can appear for trivial phases in any odd spatial dimension. We then discuss phases with topological entanglement entropy (TEE) that cannot be obtained by adding local contributions. We find that in three dimensions there is just one type of TEE, as in two dimensions, that depends linearly on the number of connected components of the boundary (the `zeroth Betti number'). In D > 3 dimensions, new types of TEE appear which depend on the higher Betti numbers of the boundary manifold. We construct generalized toric code models that exhibit these TEEs and discuss ways to extract TEE in D >=3.
△ Less
Submitted 19 August, 2011;
originally announced August 2011.
-
Quantized Response and Topology of Insulators with Inversion Symmetry
Authors:
Ari M. Turner,
Yi Zhang,
Roger S. K. Mong,
Ashvin Vishwanath
Abstract:
We study three dimensional insulators with inversion symmetry, in which other point group symmetries, such as time reversal, are generically absent. Their band topology is found to be classified by the parities of occupied states at time reversal invariant momenta (TRIM parities), and by three Chern numbers. The TRIM parities of any insulator must satisfy a constraint: their product must be +1. Th…
▽ More
We study three dimensional insulators with inversion symmetry, in which other point group symmetries, such as time reversal, are generically absent. Their band topology is found to be classified by the parities of occupied states at time reversal invariant momenta (TRIM parities), and by three Chern numbers. The TRIM parities of any insulator must satisfy a constraint: their product must be +1. The TRIM parities also constrain the Chern numbers modulo two. When the Chern numbers vanish, a magneto-electric response parameterized by "theta" is defined and is quantized to theta= 0, 2pi. Its value is entirely determined by the TRIM parities. These results may be useful in the search for magnetic topological insulators with large theta. A classification of inversion symmetric insulators is also given for general dimensions. An alternate geometrical derivation of our results is obtained by using the entanglement spectrum of the ground state wave-function.
△ Less
Submitted 3 December, 2010; v1 submitted 20 October, 2010;
originally announced October 2010.
-
Topological Phases of One-Dimensional Fermions: An Entanglement Point of View
Authors:
Ari M. Turner,
Frank Pollmann,
Erez Berg
Abstract:
The effect of interactions on topological insulators and superconductors remains, to a large extent, an open problem. Here, we describe a framework for classifying phases of one-dimensional interacting fermions, focusing on spinless fermions with time-reversal symmetry and particle number parity conservation, using concepts of entanglement. In agreement with an example presented by Fidkowski \emph…
▽ More
The effect of interactions on topological insulators and superconductors remains, to a large extent, an open problem. Here, we describe a framework for classifying phases of one-dimensional interacting fermions, focusing on spinless fermions with time-reversal symmetry and particle number parity conservation, using concepts of entanglement. In agreement with an example presented by Fidkowski \emph{et. al.} (Phys. Rev. B 81, 134509 (2010)), we find that in the presence of interactions there are only eight distinct phases, which obey a $\mathbb{Z}_8$ group structure. This is in contrast to the $\mathbb{Z}$ classification in the non-interacting case. Each of these eight phases is characterized by a unique set of bulk invariants, related to the transformation laws of its entanglement (Schmidt) eigenstates under symmetry operations, and has a characteristic degeneracy of its entanglement levels. If translational symmetry is present, the number of distinct phases increases to 16.
△ Less
Submitted 19 March, 2012; v1 submitted 25 August, 2010;
originally announced August 2010.
-
Orbital magnetoelectric coupling in band insulators
Authors:
Andrew M. Essin,
Ari M. Turner,
Joel E. Moore,
David Vanderbilt
Abstract:
Magnetoelectric responses are a fundamental characteristic of materials that break time-reversal and inversion symmetries (notably multiferroics) and, remarkably, of "topological insulators" in which those symmetries are unbroken. Previous work has shown how to compute spin and lattice contributions to the magnetoelectric tensor. Here we solve the problem of orbital contributions by computing the…
▽ More
Magnetoelectric responses are a fundamental characteristic of materials that break time-reversal and inversion symmetries (notably multiferroics) and, remarkably, of "topological insulators" in which those symmetries are unbroken. Previous work has shown how to compute spin and lattice contributions to the magnetoelectric tensor. Here we solve the problem of orbital contributions by computing the frozen-lattice electronic polarization induced by a magnetic field. One part of this response (the "Chern-Simons term") can appear even in time-reversal-symmetric materials and has been previously shown to be quantized in topological insulators. In general materials there are additional orbital contributions to all parts of the magnetoelectric tensor; these vanish in topological insulators by symmetry and also vanish in several simplified models without time-reversal and inversion those magnetoelectric couplings were studied before. We give two derivations of the response formula, one based on a uniform magnetic field and one based on extrapolation of a long-wavelength magnetic field, and discuss some of the consequences of this formula.
△ Less
Submitted 16 April, 2010; v1 submitted 1 February, 2010;
originally announced February 2010.
-
Entanglement spectrum of a topological phase in one dimension
Authors:
Frank Pollmann,
Erez Berg,
Ari M. Turner,
Masaki Oshikawa
Abstract:
We show that the Haldane phase of S=1 chains is characterized by a double degeneracy of the entanglement spectrum. The degeneracy is protected by a set of symmetries (either the dihedral group of $π$-rotations about two orthogonal axes, time-reversal symmetry, or bond centered inversion symmetry), and cannot be lifted unless either a phase boundary to another, "topologically trivial", phase is cro…
▽ More
We show that the Haldane phase of S=1 chains is characterized by a double degeneracy of the entanglement spectrum. The degeneracy is protected by a set of symmetries (either the dihedral group of $π$-rotations about two orthogonal axes, time-reversal symmetry, or bond centered inversion symmetry), and cannot be lifted unless either a phase boundary to another, "topologically trivial", phase is crossed, or the symmetry is broken. More generally, these results offer a scheme to classify gapped phases of one dimensional systems. Physically, the degeneracy of the entanglement spectrum can be observed by adiabatically weakening a bond to zero, which leaves the two disconnected halves of the system in a finitely entangled state.
△ Less
Submitted 22 April, 2010; v1 submitted 9 October, 2009;
originally announced October 2009.
-
Symmetry protection of topological order in one-dimensional quantum spin systems
Authors:
Frank Pollmann,
Erez Berg,
Ari M. Turner,
Masaki Oshikawa
Abstract:
We discuss the characterization and stability of the Haldane phase in integer spin chains on the basis of simple, physical arguments. We find that an odd-$S$ Haldane phase is a topologically non-trivial phase which is protected by any one of the following three global symmetries: (i) the dihedral group of $π$-rotations about $x,y$ and $z$ axes; (ii) time-reversal symmetry…
▽ More
We discuss the characterization and stability of the Haldane phase in integer spin chains on the basis of simple, physical arguments. We find that an odd-$S$ Haldane phase is a topologically non-trivial phase which is protected by any one of the following three global symmetries: (i) the dihedral group of $π$-rotations about $x,y$ and $z$ axes; (ii) time-reversal symmetry $S^{x,y,z} \rightarrow - S^{x,y,z}$; (iii) link inversion symmetry (reflection about a bond center), consistently with previous results [Phys. Rev. B \textbf{81}, 064439 (2010)]. On the other hand, an even-$S$ Haldane phase is not topologically protected (i.e., it is indistinct from a trivial, site-factorizable phase). We show some numerical evidence that supports these claims, using concrete examples.
△ Less
Submitted 21 March, 2012; v1 submitted 22 September, 2009;
originally announced September 2009.
-
Band Topology of Insulators via the Entanglement Spectrum
Authors:
Ari M. Turner,
Yi Zhang,
Ashvin Vishwanath
Abstract:
How do we uniquely identify a quantum phase, given its ground state wave-function? This is a key question for many body theory especially when we consider phases like topological insulators, that share the same symmetry but differ at the level of topology. The entanglement spectrum has been proposed as a ground state property that captures characteristic edge excitations. Here we study the entan…
▽ More
How do we uniquely identify a quantum phase, given its ground state wave-function? This is a key question for many body theory especially when we consider phases like topological insulators, that share the same symmetry but differ at the level of topology. The entanglement spectrum has been proposed as a ground state property that captures characteristic edge excitations. Here we study the entanglement spectrum for topological band insulators. We first show that insulators with topological surface states will necessarily also have protected modes in the entanglement spectrum. Surprisingly, however, the converse is not true. Protected entanglement modes can also appear for insulators without physical surface states, in which case they capture a more elusive property. This is illustrated by considering insulators with only inversion symmetry. Inversion is shown to act in an unusual way, as an antiunitary operator, on the entanglement spectrum, leading to this protection. The entanglement degeneracies indicate a variety of different phases in inversion symmetric insulators, and these phases are argued to be robust to the introduction of interactions.
△ Less
Submitted 9 February, 2010; v1 submitted 17 September, 2009;
originally announced September 2009.
-
Kinetic Magnetism and Orbital Order in Iron Telluride
Authors:
Ari M. Turner,
Fa Wang,
Ashvin Vishwanath
Abstract:
Iron telluride (FeTe), a relative of the iron based high temperature superconductors, displays unusual magnetic order and structural transitions. Here we explore the idea that strong correlations may play an important role in these materials. We argue that the unusual orders observed in FeTe can be understood from a picture of correlated local moments with orbital degeneracy, coupled to a small…
▽ More
Iron telluride (FeTe), a relative of the iron based high temperature superconductors, displays unusual magnetic order and structural transitions. Here we explore the idea that strong correlations may play an important role in these materials. We argue that the unusual orders observed in FeTe can be understood from a picture of correlated local moments with orbital degeneracy, coupled to a small density of itinerant electrons. A component of the structural transition is attributed to orbital, rather than magnetic ordering, introducing a strongly anisotropic character to the system along the diagonal directions of the iron lattice. Double exchange interactions couple the diagonal chains leading to the observed ordering wavevector. The incommensurate order in samples with excess iron arises from electron do** in this scenario. The strong anisotropy of physical properties in the ordered phase should be detectable by transport in single domains. Predictions for ARPES, inelastic neutron scattering and hole/electron do** studies are also made.
△ Less
Submitted 22 May, 2009;
originally announced May 2009.
-
The Mass of a Spin Vortex in a Bose-Einstein Condensate
Authors:
Ari M. Turner
Abstract:
In contrast to charge vortices in a superfluid, spin vortices in a ferromagnetic condensate move inertially (if the condensate has zero magnetization along an axis). The mass of spin vortices depends on the spin-dependent interactions, and can be measured as a part of experiments on how spin vortices orbit one another. For Rb87 in a 1 micron thick trap m_v is about 10^-21 kg.
In contrast to charge vortices in a superfluid, spin vortices in a ferromagnetic condensate move inertially (if the condensate has zero magnetization along an axis). The mass of spin vortices depends on the spin-dependent interactions, and can be measured as a part of experiments on how spin vortices orbit one another. For Rb87 in a 1 micron thick trap m_v is about 10^-21 kg.
△ Less
Submitted 6 March, 2009; v1 submitted 20 February, 2009;
originally announced February 2009.
-
Vortex Molecules in Spinor Condensates
Authors:
Ari M. Turner,
Eugene Demler
Abstract:
Condensates of atoms with spins can have vortices of several types; these are related to the symmetry group of the atoms' ground state. We discuss how, when a condensate is placed in a small magnetic field that breaks the spin symmetry, these vortices may form bound states. Using symmetry classification of vortex-charge and rough estimates for vortex interactions, one can show that some configur…
▽ More
Condensates of atoms with spins can have vortices of several types; these are related to the symmetry group of the atoms' ground state. We discuss how, when a condensate is placed in a small magnetic field that breaks the spin symmetry, these vortices may form bound states. Using symmetry classification of vortex-charge and rough estimates for vortex interactions, one can show that some configurations that are stable at zero temperature can decay at finite temperatures by crossing over energy barriers. Our focus is cyclic spin 2 condensates, which have tetrahedral symmetry.
△ Less
Submitted 30 July, 2008;
originally announced July 2008.
-
Vortices in Superfluid Films on Curved Surfaces
Authors:
Ari M. Turner,
Vincenzo Vitelli,
David R. Nelson
Abstract:
We present a systematic study of how vortices in superfluid films interact with the spatially varying Gaussian curvature of the underlying substrate. The Gaussian curvature acts as a source for a geometric potential that attracts (repels) vortices towards regions of negative (positive) Gaussian curvature independently of the sign of their topological charge. Various experimental tests involving…
▽ More
We present a systematic study of how vortices in superfluid films interact with the spatially varying Gaussian curvature of the underlying substrate. The Gaussian curvature acts as a source for a geometric potential that attracts (repels) vortices towards regions of negative (positive) Gaussian curvature independently of the sign of their topological charge. Various experimental tests involving rotating superfluid films and vortex pinning are first discussed for films coating gently curved substrates that can be treated in perturbation theory from flatness. An estimate of the experimental regimes of interest is obtained by comparing the strength of the geometrical forces to the vortex pinning induced by the varying thickness of the film which is in turn caused by capillary effects and gravity. We then present a non-perturbative technique based on conformal map**s that leads an exact solution for the geometric potential as well as the geometric correction to the interaction between vortices. The conformal map** approach is illustrated by means of explicit calculations of the geometric effects encountered in the study of some strongly curved surfaces and by deriving universal bounds on their strength.
△ Less
Submitted 2 July, 2008;
originally announced July 2008.
-
The HST Key Project on the Extragalactic Distance Scale XVII. The Cepheid Distance to NGC 4725
Authors:
Brad K. Gibson,
Shaun M. G. Hughes,
Peter B. Stetson,
Wendy L. Freedman,
Robert C. Kennicutt, Jr.,
Jeremy R. Mould,
Fabio Bresolin,
Laura Ferrarese,
Holland C. Ford,
John A. Graham,
Mingsheng Han,
Paul Harding,
John G. Hoessel,
John P. Huchra,
Garth D. Illingworth,
Daniel D. Kelson,
Lucas M. Macri,
Barry F. Madore,
Randy L. Phelps,
Charles F. Prosser,
Abhijit Saha,
Shoko Sakai,
Kim M. Sebo,
Nancy A. Silbermann,
Anne M. Turner
Abstract:
The distance to NGC 4725 has been derived from Cepheid variables, as part of the Hubble Space Telescope Key Project on the Extragalactic Distance Scale. Thirteen F555W (V) and four F814W (I) epochs of cosmic-ray-split Wide Field and Planetary Camera 2 observations were obtained. Twenty Cepheids were discovered, with periods ranging from 12 to 49 days. Adopting a Large Magellanic Cloud distance m…
▽ More
The distance to NGC 4725 has been derived from Cepheid variables, as part of the Hubble Space Telescope Key Project on the Extragalactic Distance Scale. Thirteen F555W (V) and four F814W (I) epochs of cosmic-ray-split Wide Field and Planetary Camera 2 observations were obtained. Twenty Cepheids were discovered, with periods ranging from 12 to 49 days. Adopting a Large Magellanic Cloud distance modulus and extinction of 18.50+/-0.10 mag and E(V-I)=0.13 mag, respectively, a true reddening-corrected distance modulus (based on an analysis employing the ALLFRAME software package) of 30.50 +/- 0.16 (random) +/- 0.17 (systematic) mag was determined for NGC 4725. The corresponding of distance of 12.6 +/- 1.0 (random) +/- 1.0 (systematic) Mpc is in excellent agreement with that found with an independent analysis based upon the DoPHOT photometry package. With a foreground reddening of only E(V-I)=0.02, the inferred intrinsic reddening of this field in NGC 4725, E(V-I)=0.19, makes it one of the most highly-reddened, encountered by the HST Key Project, to date.
△ Less
Submitted 1 October, 1998;
originally announced October 1998.