-
Weyl semimetallic state with antiferromagnetic order in Rashba-Hubbard model
Authors:
Aastha Jain,
Garima Goyal,
Dheeraj Kumar Singh
Abstract:
We study the phase diagram of Rashba-Hubbard model by employing the Hartree-Fock meanfield theory, and thereby establish the existence of an antiferromagnetically ordered Weyl semimetallic state with in-plane magnetic moments. This phase is found to be sandwiched in between the antiferromagnetic insulator and Rashba metal in the interaction vs spin-orbit coupling phase diagram. The antiferromagnet…
▽ More
We study the phase diagram of Rashba-Hubbard model by employing the Hartree-Fock meanfield theory, and thereby establish the existence of an antiferromagnetically ordered Weyl semimetallic state with in-plane magnetic moments. This phase is found to be sandwiched in between the antiferromagnetic insulator and Rashba metal in the interaction vs spin-orbit coupling phase diagram. The antiferromagnetically-ordered topological semimetallic state exists in the presence of combined time-reversal and inversion symmetry though individually both are broken. The study of the static magnetic susceptibility indicates the robustness of the antiferromagnetic order within a realistic range of interaction and spin-orbit coupling parameters. In addition to the edge states associated with the Weyl points, we also investigate the spin-resolved quasiparticle interference, which provides important insight into the possible spin texture of the bands especially in the vicinity of Weyl points.
△ Less
Submitted 3 February, 2024;
originally announced February 2024.
-
Antiferromagnetically ordered Dirac semimetal in Hubbard model with spin-orbit coupling
Authors:
Garima Goyal,
Dheeraj Kumar Singh
Abstract:
We examine the possible existence of Dirac semimetal with magnetic order in a two-dimensional system with a nonsymmorphic symmetry by using the Hartree-Fock mean-field theory within the Hubbard model. We locate the region in the second-neighbor spin-orbit coupling vs Hubbard interaction phase diagram, where such a state is stabilized. The edge states for the ribbons along two orthogonal directions…
▽ More
We examine the possible existence of Dirac semimetal with magnetic order in a two-dimensional system with a nonsymmorphic symmetry by using the Hartree-Fock mean-field theory within the Hubbard model. We locate the region in the second-neighbor spin-orbit coupling vs Hubbard interaction phase diagram, where such a state is stabilized. The edge states for the ribbons along two orthogonal directions concerning the orientation of in-plane magnetic moments are obtained. Finally, the effect of the in-plane magnetic field, which results in the stabilization of the Weyl semimetallic state, and the nature of the edge states corresponding to the Weyl semimetallic state for ribbon geometries are also explored.
△ Less
Submitted 7 January, 2024; v1 submitted 29 March, 2023;
originally announced March 2023.
-
Temperature dependence of quasiparticle interference in $d$-wave superconductors
Authors:
Harun Al Rashid,
Garima Goyal,
Alireza Akbari,
Dheeraj Kumar Singh
Abstract:
We investigate the temperature dependence of quasiparticle interference in the high $T_c$-cuprates using an Exact-Diagonalization + Monte-Carlo based scheme to simulate the $d$-wave superconducting order parameter. The quasiparticle interference patterns have features largely resulting from the scattering vectors of the octet model at lower temperature. Our findings suggest that the features of qu…
▽ More
We investigate the temperature dependence of quasiparticle interference in the high $T_c$-cuprates using an Exact-Diagonalization + Monte-Carlo based scheme to simulate the $d$-wave superconducting order parameter. The quasiparticle interference patterns have features largely resulting from the scattering vectors of the octet model at lower temperature. Our findings suggest that the features of quasiparticle interference in the pseudogap region of the phase diagram are also dominated by the set of scattering vectors belonging to the octet model because of the persisting antinodal gap beyond the superconducting transition $T_c$. However, beyond a temperature when the antinodal gap becomes very small, a set of scattering vectors different from those belonging to the octet model are responsible for the quasiparticle interference patterns. With a rise in temperature, the patterns are increasingly broadened.
△ Less
Submitted 15 December, 2022; v1 submitted 19 October, 2022;
originally announced October 2022.
-
Semimetallic spin-density wave state in iron pnictides
Authors:
Garima Goyal,
Dheeraj Kumar Singh
Abstract:
We examine the existence of semimetallic spin-density wave states in iron pnictides. In the experimentally observed metallic spin-density wave state, the symmetry-protected Dirac cones are located away from the Fermi surface giving rise to tiny pockets and there are also additional Fermi pockets such as one around $Γ$. We find that the location of a pair of Dirac points with respect to the Fermi s…
▽ More
We examine the existence of semimetallic spin-density wave states in iron pnictides. In the experimentally observed metallic spin-density wave state, the symmetry-protected Dirac cones are located away from the Fermi surface giving rise to tiny pockets and there are also additional Fermi pockets such as one around $Γ$. We find that the location of a pair of Dirac points with respect to the Fermi surface exhibits significant sensitivity to the orbital splitting between the $d_{xz}$ and $d_{yz}$ orbitals. Besides, in the presence of orbital splitting, the Fermi pockets not associated with the Dirac cones, can be suppressed so that a semimetallic spin-density wave state can be realized. We explain these finding in terms of difference in the slopes and orbital contents of the bands which form the Dirac cone, and obtain the necessary conditions dependent on these two and other parameters for the coexisting Dirac semimetallic and spin-density wave states. Additionally, the topologically protected edge states are studied in the ribbon geometry when the same are oriented either along $x$ or $y$ axes.
△ Less
Submitted 9 July, 2022;
originally announced July 2022.
-
Hund's coupling and electronic anisotropy in the spin-density wave state of iron pnictides
Authors:
Garima Goyal,
Dheeraj Kumar Singh
Abstract:
In the multiband systems, Hund's coupling ($J$) plays a significant role in the spin and charge excitations. We study the dependence of electronic anisotropy on $J$ in terms of Drude-weight along different directions as well as the orbital order in the four-fold symmetry broken spin-density wave state of iron pnictides. A robust behavior of the Drude-weight anisotropy within a small window around…
▽ More
In the multiband systems, Hund's coupling ($J$) plays a significant role in the spin and charge excitations. We study the dependence of electronic anisotropy on $J$ in terms of Drude-weight along different directions as well as the orbital order in the four-fold symmetry broken spin-density wave state of iron pnictides. A robust behavior of the Drude-weight anisotropy within a small window around $J \sim 0.25U$ with $U$ as intraorbital Coulomb interaction is described in terms of orbital-weight distribution along the reconstructed Fermi surfaces. We also find that the ferro-orbital order increases with $J$ in the widely accepted regime for the latter, which is explained as a consequence of rising exchange field with an increase in magnetization.
△ Less
Submitted 24 January, 2022;
originally announced January 2022.
-
AI Powered Compiler Techniques for DL Code Optimization
Authors:
Sanket Tavarageri,
Gagandeep Goyal,
Sasikanth Avancha,
Bharat Kaul,
Ramakrishna Upadrasta
Abstract:
Creating high performance implementations of deep learning primitives on CPUs is a challenging task. Multiple considerations including multi-level cache hierarchy, and wide SIMD units of CPU platforms influence the choice of program transformations to apply for performance optimization. In this paper, we present machine learning powered compiler techniques to optimize loop nests. We take a two-pro…
▽ More
Creating high performance implementations of deep learning primitives on CPUs is a challenging task. Multiple considerations including multi-level cache hierarchy, and wide SIMD units of CPU platforms influence the choice of program transformations to apply for performance optimization. In this paper, we present machine learning powered compiler techniques to optimize loop nests. We take a two-pronged approach to code optimization: We first apply high level optimizations to optimize the code to take optimal advantage of the cache memories. Then, we perform low level, target-specific optimizations to effectively vectorize the code to run well on the SIMD units of the machine. For high level optimizations, we use polyhedral compilation techniques and deep learning approaches. For low level optimization, we use a target specific code generator that generates code using vector intrinsics and Reinforcement Learning (RL) techniques to find the optimal parameters for the code generator. We perform experimental evaluation of the developed techniques on various matrix multiplications that occur in popular deep learning workloads. The experimental results show that the compiler techniques presented in the paper achieve 7.6X and 8.2X speed-ups over a baseline for sequential and parallel runs respectively.
△ Less
Submitted 12 April, 2021;
originally announced April 2021.
-
PolyDL: Polyhedral Optimizations for Creation of High Performance DL primitives
Authors:
Sanket Tavarageri,
Alexander Heinecke,
Sasikanth Avancha,
Gagandeep Goyal,
Ramakrishna Upadrasta,
Bharat Kaul
Abstract:
Deep Neural Networks (DNNs) have revolutionized many aspects of our lives. The use of DNNs is becoming ubiquitous including in softwares for image recognition, speech recognition, speech synthesis, language translation, to name a few. he training of DNN architectures however is computationally expensive. Once the model is created, its use in the intended application - the inference task, is comput…
▽ More
Deep Neural Networks (DNNs) have revolutionized many aspects of our lives. The use of DNNs is becoming ubiquitous including in softwares for image recognition, speech recognition, speech synthesis, language translation, to name a few. he training of DNN architectures however is computationally expensive. Once the model is created, its use in the intended application - the inference task, is computationally heavy too and the inference needs to be fast for real time use. For obtaining high performance today, the code of Deep Learning (DL) primitives optimized for specific architectures by expert programmers exposed via libraries is the norm. However, given the constant emergence of new DNN architectures, creating hand optimized code is expensive, slow and is not scalable.
To address this performance-productivity challenge, in this paper we present compiler algorithms to automatically generate high performance implementations of DL primitives that closely match the performance of hand optimized libraries. We develop novel data reuse analysis algorithms using the polyhedral model to derive efficient execution schedules automatically. In addition, because most DL primitives use some variant of matrix multiplication at their core, we develop a flexible framework where it is possible to plug in library implementations of the same in lieu of a subset of the loops. We show that such a hybrid compiler plus a minimal library-use approach results in state-of-the-art performance. We develop compiler algorithms to also perform operator fusions that reduce data movement through the memory hierarchy of the computer system.
△ Less
Submitted 17 November, 2020; v1 submitted 2 June, 2020;
originally announced June 2020.
-
PolyScientist: Automatic Loop Transformations Combined with Microkernels for Optimization of Deep Learning Primitives
Authors:
Sanket Tavarageri,
Alexander Heinecke,
Sasikanth Avancha,
Gagandeep Goyal,
Ramakrishna Upadrasta,
Bharat Kaul
Abstract:
At the heart of deep learning training and inferencing are computationally intensive primitives such as convolutions which form the building blocks of deep neural networks. Researchers have taken two distinct approaches to creating high performance implementations of deep learning kernels, namely, 1) library development exemplified by Intel MKL-DNN for CPUs, 2) automatic compilation represented by…
▽ More
At the heart of deep learning training and inferencing are computationally intensive primitives such as convolutions which form the building blocks of deep neural networks. Researchers have taken two distinct approaches to creating high performance implementations of deep learning kernels, namely, 1) library development exemplified by Intel MKL-DNN for CPUs, 2) automatic compilation represented by the TensorFlow XLA compiler. The two approaches have their drawbacks: even though a custom built library can deliver very good performance, the cost and time of development of the library can be high. Automatic compilation of kernels is attractive but in practice, till date, automatically generated implementations lag expert coded kernels in performance by orders of magnitude.
In this paper, we develop a hybrid solution to the development of deep learning kernels that achieves the best of both worlds: the expert coded microkernels are utilized for the innermost loops of kernels and we use the advanced polyhedral technology to automatically tune the outer loops for performance. We design a novel polyhedral model based data reuse algorithm to optimize the outer loops of the kernel. Through experimental evaluation on an important class of deep learning primitives namely convolutions, we demonstrate that the approach we develop attains the same levels of performance as Intel MKL-DNN, a hand coded deep learning library.
△ Less
Submitted 6 February, 2020;
originally announced February 2020.
-
The role of ego vision in view-invariant action recognition
Authors:
Gaurvi Goyal,
Nicoletta Noceti,
Francesca Odone,
Alessandra Sciutti
Abstract:
Analysis and interpretation of egocentric video data is becoming more and more important with the increasing availability and use of wearable cameras. Exploring and fully understanding affinities and differences between ego and allo (or third-person) vision is paramount for the design of effective methods to process, analyse and interpret egocentric data. In addition, a deeper understanding of ego…
▽ More
Analysis and interpretation of egocentric video data is becoming more and more important with the increasing availability and use of wearable cameras. Exploring and fully understanding affinities and differences between ego and allo (or third-person) vision is paramount for the design of effective methods to process, analyse and interpret egocentric data. In addition, a deeper understanding of ego-vision and its peculiarities may enable new research perspectives in which first person viewpoints can act either as a mean for easily acquiring large amounts of data to be employed in general-purpose recognition systems, and as a challenging test-bed to assess the usability of techniques specifically tailored to deal with allocentric vision on more challenging settings. Our work, with an eye to cognitive science findings, leverages transfer learning in Convolutional Neural Networks to demonstrate capabilities and limitations of an implicitly learnt view-invariant representation in the specific case of action recognition.
△ Less
Submitted 10 June, 2019;
originally announced June 2019.
-
The effect of gait on swimming in viscoelastic fluids
Authors:
Gwynn J. Elfring,
Gaurav Goyal
Abstract:
In this paper, we give formulas for the swimming of simplified two-dimensional bodies in complex fluids using the reciprocal theorem. By way of these formulas we calculate the swimming velocity due to small-amplitude deformations on the simplest of these bodies, a two-dimensional sheet, to explore general conditions on the swimming gait under which the sheet may move faster, or slower, in a viscoe…
▽ More
In this paper, we give formulas for the swimming of simplified two-dimensional bodies in complex fluids using the reciprocal theorem. By way of these formulas we calculate the swimming velocity due to small-amplitude deformations on the simplest of these bodies, a two-dimensional sheet, to explore general conditions on the swimming gait under which the sheet may move faster, or slower, in a viscoelastic fluid compared to a Newtonian fluid. We show that in general, for small amplitude deformations, a speed increase can only be realized by multiple deformation modes in contrast to slip flows. Additionally, we show that a change in swimming speed is directly due to a change in thrust generated by the swimmer.
△ Less
Submitted 19 April, 2016; v1 submitted 19 November, 2015;
originally announced November 2015.
-
Neuronal micro-culture engineering by microchannel devices of cellular scale dimensions
Authors:
Gaurav Goyal,
Yoonkey Nam
Abstract:
Purpose: The purpose of the current study was to investigate the effect of microchannel geometry on neuronal cultures and to maintain these cultures for long period of time (over several weeks) inside the closed microchannels of cellular scale dimensions.
Methods: The primary hippocampal neurons from E-18 rat were cultured inside the closed polydimethylsiloxane (PDMS) microchannels of varying si…
▽ More
Purpose: The purpose of the current study was to investigate the effect of microchannel geometry on neuronal cultures and to maintain these cultures for long period of time (over several weeks) inside the closed microchannels of cellular scale dimensions.
Methods: The primary hippocampal neurons from E-18 rat were cultured inside the closed polydimethylsiloxane (PDMS) microchannels of varying sizes. The effect of the channel geometry on the spatial and the temporal variations in the neural microenvironment was investigated by studying neural maturation and variation in the media osmolality respectively. The cultures were maintained for longer time spans by PDMS device pretreatment, control on media evaporation (by using hydrophobic ethylene propylene membrane) and an effective culture maintenance protocol. Further, the devices were integrated with the planar microelectrode arrays (MEA) to record spontaneous electrical activity.
Results: A direct influence of channel geometry on neuron maturation was observed with cells in smaller channels maturing faster. The temporal variation in the microenvironment was caused by several fold increase in osmolality within 2-3 days due to rapid media evaporation. With our culture methodology, neurons were maintained in the closed channels as small as 50 microns in height and width for over 1 month in serum free media condition and the time varying spontaneous electrical activity was measured for up to 5 weeks using the MEA.
Conclusions: The understanding of the effect of the culture scale on cellular microenvironment and such long-term culture maintenance will be helpful in studying neuronal tissue development; therapeutic drug screening; and for network level neuronal analysis.
△ Less
Submitted 5 February, 2015;
originally announced February 2015.
