arXiv subjects

Miao Liu

Publications and source records attributed to Miao Liu.

At least 127 records · Page 7Linked to original sources

arXiv2022

Screening promising CsV3Sb5-like kagome materials from systematic first-principles evaluation

CsV3Sb5 kagome lattice holds the promise for manifesting electron correlation, topology and superconducting. However, by far only three CsV3Sb5-like kagome materials have been experimentally spotted. In this work, we enlarge this family of materials to 1386 compounds via element species substitution, and the further screening process suggests that 28 promising candidates have superior thermodynamic stability, hence they are highly likely to be synthesized. Moreover, these compounds possess several identical electronic structures, and can be categorized into five non-magnetic and three magnetic groups accordingly. It is our hope that this work can greatly expand the viable phase space of the CsV3Sb5-like materials for investigating or tuning the novel quantum phenomena in kagome lattice.

cond-mat.mtrl-sci↗

arXiv2022

Unconventional Materials: the mismatch between electronic charge centers and atomic positions

The complete band representations (BRs) have been constructed in the work of topological quantum chemistry. Each BR is expressed by either a localized orbital at a Wyckoff site in real space, or by a set of irreducible representations in momentum space. In this work, we define unconventional materials with a common feature of the mismatch between average electronic centers and atomic positions. They can be effectively diagnosed as whose occupied bands can be expressed as a sum of elementary BRs (eBRs), but not a sum of atomic-orbital-induced BRs (aBRs). The existence of an essential BR at an empty site is described by nonzero real-space invariants (RSIs). The "valence" states can be derived by the aBR decomposition, and unconventional materials are supposed to have an uncompensated total "valence" state. The high-throughput screening for unconventional materials has been performed through the first-principles calculations. We have discovered 423 unconventional compounds, including thermoelectronic materials, higher-order topological insulators, electrides, hydrogen storage materials, hydrogen evolution reaction electrocatalysts, electrodes, and superconductors. The diversity of these interesting properties and applications would be widely studied in the future.

cond-mat.mtrl-sci↗

arXiv2021

4D Human Body Capture from Egocentric Video via 3D Scene Grounding

We introduce a novel task of reconstructing a time series of second-person 3D human body meshes from monocular egocentric videos. The unique viewpoint and rapid embodied camera motion of egocentric videos raise additional technical barriers for human body capture. To address those challenges, we propose a simple yet effective optimization-based approach that leverages 2D observations of the entire video sequence and human-scene interaction constraint to estimate second-person human poses, shapes, and global motion that are grounded on the 3D environment captured from the egocentric view. We conduct detailed ablation studies to validate our design choice. Moreover, we compare our method with the previous state-of-the-art method on human motion capture from monocular video, and show that our method estimates more accurate human-body poses and shapes under the challenging egocentric setting. In addition, we demonstrate that our approach produces more realistic human-scene interaction.

cs.CV↗

arXiv2021

A universal model for the formation energy prediction of inorganic compounds

Harnessing the recent advance in data science and materials science, it is feasible today to build predictive models for materials properties. In this study, we employ the data of high-throughput quantum mechanics calculations based on 170,714 inorganic crystalline compounds to train a machine learning model for formation energy prediction. Different from the previous work, our model reaches a fairly good predictive ability (R2=0.982 and MAE=0.07 eVatom-1, DenseNet model) and meanwhile can be universally applied to the large phase space of inorganic materials. The improvement comes from several effective structure-dependent descriptors that are proposed to take the information of electronegativity and structure into account. This model can provide a useful tool to search for new materials in a vast phase space in a fast and cost-effective manner.

cond-mat.mtrl-sci↗

arXiv2021

A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement Learning

A fundamental challenge in multiagent reinforcement learning is to learn beneficial behaviors in a shared environment with other simultaneously learning agents. In particular, each agent perceives the environment as effectively non-stationary due to the changing policies of other agents. Moreover, each agent is itself constantly learning, leading to natural non-stationarity in the distribution of experiences encountered. In this paper, we propose a novel meta-multiagent policy gradient theorem that directly accounts for the non-stationary policy dynamics inherent to multiagent learning settings. This is achieved by modeling our gradient updates to consider both an agent's own non-stationary policy dynamics and the non-stationary policy dynamics of other agents in the environment. We show that our theoretically grounded approach provides a general solution to the multiagent learning problem, which inherently comprises all key aspects of previous state of the art approaches on this topic. We test our method on a diverse suite of multiagent benchmarks and demonstrate a more efficient ability to adapt to new agents as they learn than baseline methods across the full spectrum of mixed incentive, competitive, and cooperative domains.

cs.LG↗

arXiv2020

Consolidation via Policy Information Regularization in Deep RL for Multi-Agent Games

This paper introduces an information-theoretic constraint on learned policy complexity in the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) reinforcement learning algorithm. Previous research with a related approach in continuous control experiments suggests that this method favors learning policies that are more robust to changing environment dynamics. The multi-agent game setting naturally requires this type of robustness, as other agents' policies change throughout learning, introducing a nonstationary environment. For this reason, recent methods in continual learning are compared to our approach, termed Capacity-Limited MADDPG. Results from experimentation in multi-agent cooperative and competitive tasks demonstrate that the capacity-limited approach is a good candidate for improving learning performance in these environments.

cs.AI↗

arXiv2020

In the Eye of the Beholder: Gaze and Actions in First Person Video

We address the task of jointly determining what a person is doing and where they are looking based on the analysis of video captured by a headworn camera. To facilitate our research, we first introduce the EGTEA Gaze+ dataset. Our dataset comes with videos, gaze tracking data, hand masks and action annotations, thereby providing the most comprehensive benchmark for First Person Vision (FPV). Moving beyond the dataset, we propose a novel deep model for joint gaze estimation and action recognition in FPV. Our method describes the participant's gaze as a probabilistic variable and models its distribution using stochastic units in a deep network. We further sample from these stochastic units, generating an attention map to guide the aggregation of visual features for action recognition. Our method is evaluated on our EGTEA Gaze+ dataset and achieves a performance level that exceeds the state-of-the-art by a significant margin. More importantly, we demonstrate that our model can be applied to larger scale FPV dataset---EPIC-Kitchens even without using gaze, offering new state-of-the-art results on FPV action recognition.

cs.CV↗

arXiv2020

Deep RL With Information Constrained Policies: Generalization in Continuous Control

Biological agents learn and act intelligently in spite of a highly limited capacity to process and store information. Many real-world problems involve continuous control, which represents a difficult task for artificial intelligence agents. In this paper we explore the potential learning advantages a natural constraint on information flow might confer onto artificial agents in continuous control tasks. We focus on the model-free reinforcement learning (RL) setting and formalize our approach in terms of an information-theoretic constraint on the complexity of learned policies. We show that our approach emerges in a principled fashion from the application of rate-distortion theory. We implement a novel Capacity-Limited Actor-Critic (CLAC) algorithm and situate it within a broader family of RL algorithms such as the Soft Actor Critic (SAC) and Mutual Information Reinforcement Learning (MIRL) algorithm. Our experiments using continuous control tasks show that compared to alternative approaches, CLAC offers improvements in generalization between training and modified test environments. This is achieved in the CLAC model while displaying the high sample efficiency of similar methods.

cs.LG↗

arXiv2020

Attention Distillation for Learning Video Representations

We address the challenging problem of learning motion representations using deep models for video recognition. To this end, we make use of attention modules that learn to highlight regions in the video and aggregate features for recognition. Specifically, we propose to leverage output attention maps as a vehicle to transfer the learned representation from a motion (flow) network to an RGB network. We systematically study the design of attention modules, and develop a novel method for attention distillation. Our method is evaluated on major action benchmarks, and consistently improves the performance of the baseline RGB network by a significant margin. Moreover, we demonstrate that our attention maps can leverage motion cues in learning to identify the location of actions in video frames. We believe our method provides a step towards learning motion-aware representations in deep models. Our project page is available at https://aptx4869lm.github.io/AttentionDistillation/

cs.CV↗

arXiv2020

Forecasting Human-Object Interaction: Joint Prediction of Motor Attention and Actions in First Person Video

We address the challenging task of anticipating human-object interaction in first person videos. Most existing methods ignore how the camera wearer interacts with the objects, or simply consider body motion as a separate modality. In contrast, we observe that the international hand movement reveals critical information about the future activity. Motivated by this, we adopt intentional hand movement as a future representation and propose a novel deep network that jointly models and predicts the egocentric hand motion, interaction hotspots and future action. Specifically, we consider the future hand motion as the motor attention, and model this attention using latent variables in our deep model. The predicted motor attention is further used to characterise the discriminative spatial-temporal visual features for predicting actions and interaction hotspots. We present extensive experiments demonstrating the benefit of the proposed joint model. Importantly, our model produces new state-of-the-art results for action anticipation on both EGTEA Gaze+ and the EPIC-Kitchens datasets. Our project page is available at https://aptx4869lm.github.io/ForecastingHOI/

cs.CV↗

arXiv2020

Learning Hierarchical Teaching Policies for Cooperative Agents

Collective learning can be greatly enhanced when agents effectively exchange knowledge with their peers. In particular, recent work studying agents that learn to teach other teammates has demonstrated that action advising accelerates team-wide learning. However, the prior work has simplified the learning of advising policies by using simple function approximations and only considered advising with primitive (low-level) actions, limiting the scalability of learning and teaching to complex domains. This paper introduces a novel learning-to-teach framework, called hierarchical multiagent teaching (HMAT), that improves scalability to complex environments by using the deep representation for student policies and by advising with more expressive extended action sequences over multiple levels of temporal abstraction. Our empirical evaluations demonstrate that HMAT improves team-wide learning progress in large, complex domains where previous approaches fail. HMAT also learns teaching policies that can effectively transfer knowledge to different teammates with knowledge of different tasks, even when the teammates have heterogeneous action spaces.

cs.LG↗

arXiv2020

On the Role of Weight Sharing During Deep Option Learning

The options framework is a popular approach for building temporally extended actions in reinforcement learning. In particular, the option-critic architecture provides general purpose policy gradient theorems for learning actions from scratch that are extended in time. However, past work makes the key assumption that each of the components of option-critic has independent parameters. In this work we note that while this key assumption of the policy gradient theorems of option-critic holds in the tabular case, it is always violated in practice for the deep function approximation setting. We thus reconsider this assumption and consider more general extensions of option-critic and hierarchical option-critic training that optimize for the full architecture with each update. It turns out that not assuming parameter independence challenges a belief in prior work that training the policy over options can be disentangled from the dynamics of the underlying options. In fact, learning can be sped up by focusing the policy over options on states where options are actually likely to terminate. We put our new algorithms to the test in application to sample efficient learning of Atari games, and demonstrate significantly improved stability and faster convergence when learning long options.

cs.LG↗

arXiv2020

Characterization and Thermal Management of a DC Motor-Driven Resonant Actuator for Miniature Mobile Robots with Oscillating Limbs

In this paper, we characterize the performance of and develop thermal management solutions for a DC motor-driven resonant actuator developed for flapping wing micro air vehicles. The actuator, a DC micro-gearmotor connected in parallel with a torsional spring, drives reciprocal wing motion. Compared to the gearmotor alone, this design increased torque and power density by 161.1% and 666.8%, respectively, while decreasing the drawn current by 25.8%. Characterization of the actuator, isolated from nonlinear aerodynamic loading, results in standard metrics directly comparable to other actuators. The micro-motor, selected for low weight considerations, operates at high power for limited duration due to thermal effects. To predict system performance, a lumped parameter thermal circuit model was developed. Critical model parameters for this micro-motor, two orders of magnitude smaller than those previously characterized, were identified experimentally. This included the effects of variable winding resistance, bushing friction, speed-dependent forced convection, and the addition of a heatsink. The model was then used to determine a safe operation envelope for the vehicle and to design a weight-optimal heatsink. This actuator design and thermal modeling approach could be applied more generally to improve the performance of any miniature mobile robot or device with motor-driven oscillating limbs or loads.

eess.SP↗

arXiv2019

Learning Abstract Options

Building systems that autonomously create temporal abstractions from data is a key challenge in scaling learning and planning in reinforcement learning. One popular approach for addressing this challenge is the options framework (Sutton et al., 1999). However, only recently in (Bacon et al., 2017) was a policy gradient theorem derived for online learning of general purpose options in an end to end fashion. In this work, we extend previous work on this topic that only focuses on learning a two-level hierarchy including options and primitive actions to enable learning simultaneously at multiple resolutions in time. We achieve this by considering an arbitrarily deep hierarchy of options where high level temporally extended options are composed of lower level options with finer resolutions in time. We extend results from (Bacon et al., 2017) and derive policy gradient theorems for a deep hierarchy of options. Our proposed hierarchical option-critic architecture is capable of learning internal policies, termination conditions, and hierarchical compositions over options without the need for any intrinsic rewards or subgoals. Our empirical results in both discrete and continuous environments demonstrate the efficiency of our framework.

cs.LG↗

arXiv2019

Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference

Lack of performance when it comes to continual learning over non-stationary distributions of data remains a major challenge in scaling neural network learning to more human realistic settings. In this work we propose a new conceptualization of the continual learning problem in terms of a temporally symmetric trade-off between transfer and interference that can be optimized by enforcing gradient alignment across examples. We then propose a new algorithm, Meta-Experience Replay (MER), that directly exploits this view by combining experience replay with optimization based meta-learning. This method learns parameters that make interference based on future gradients less likely and transfer based on future gradients more likely. We conduct experiments across continual lifelong supervised learning benchmarks and non-stationary reinforcement learning environments demonstrating that our approach consistently outperforms recently proposed baselines for continual learning. Our experiments show that the gap between the performance of MER and baseline algorithms grows both as the environment gets more non-stationary and as the fraction of the total experiences stored gets smaller.

cs.LG↗

arXiv2019

High-throughput investigation of tunable superconductivity in FeSe films

There is an ongoing debate about the relative importance of structural change versus doping charge carriers on the mechanism of superconductivity in Fe-based materials. Elucidating this issue is a major challenge since it would require a large number of samples where structure properties or the carrier density is systematically varied. FeSe, with its structural simplicity, is an ideal platform for addressing this question. It has been demonstrated that the superconductivity in this material can be controlled through crystal lattice tuning, as well as electronic structure manipulation. Here, we apply a high-throughput methodology to FeSe to systematically delineate the interdependence of its structural and electronic properties. Using a dual-beam pulsed laser deposition, we have generated FeSe films with a marked gradient in the superconducting transition temperature (below 2 K < Tc < 12 K) across 1 cm width of the films. The Tc gradient films display ~ 1% continuous stretch and compression in the out-of-plane and in-plane lattice constants respectively, triggering the continuous enhancement of superconductivity. Combining transport and angular-resolved photoemission measurements on uniform FeSe films with tunable Tc from 3 K to 14 K, we find that the electron carrier density is intimately correlated with Tc, i.e., it increases by a factor of 6 and ultimately surpasses the almost constant hole concentration. Our transmission electron microscope and band structure calculations reveal that rather than by shifting the chemical potential, the enhanced superconductivity is linked to the selective adjustment of the dxy band dispersion across the Fermi level by means of reduced local lattice distortions. Therefore, such novel mechanism provides a key to understand discrete superconducting phases in FeSe.

cond-mat.supr-con↗

arXiv2019

Flat AgTe Honeycomb Monolayer with Topologically Nontrivial States

The intriguing properties, especially Dirac physics in graphene, have inspired the pursuit of two-dimensional materials in honeycomb structure. Here we achieved a monolayer transition metal monochalcogenide AgTe on Ag(111) by tellurization of the substrate. High-resolution scanning tunneling microscopy, combined with low-energy electron diffraction, angle-resolved photoemission spectroscopy, and density functional theory calculations, demonstrates the planar honeycomb structure of AgTe. The first principle calculations further reveal that, protected by the in-plane mirror reflection symmetry, two Dirac node-line Fermions exist in the electronic structures of free-standing AgTe when spin-orbit coupling (SOC) is ignored. While in fact the SOC leads to the gap opening, and resulting in the emergence of the topologically nontrivial quantum spin Hall edge state. Importantly, our experiments evidence the chemical stability of the monolayer AgTe in ambient conditions. It is possible to study AgTe by more ex-situ measurements and even to apply it in novel electronic devices.

cond-mat.mtrl-sci↗

arXiv2018

Evaluating the Effectiveness of Health Awareness Events by Google Search Frequency

Over two hundreds health awareness events take place in the United States in order to raise attention and educate the public about diseases. It would be informative and instructive for the organization to know the impact of these events, although such information could be difficult to measure. Here 46 events are selected and their data from 2004 to 2017 are downloaded from Google Trend(GT). We investigate whether the events effectively attract the public attention by increasing the search frequencies of certain keywords which we call queries. Three statistical methods including Transfer Function Noise modeling, Wilcoxon Rank Sum test, and Binomial inference are conducted on 46 GT data sets. Our study show that 10 health awareness events are effective with evidence of a significant increase in search frequencies in the event months, and 28 events are ineffective, with the rest being classified as unclear.

stat.AP↗