Papers made digestable
Our architecture simplifies the obstacle-perception
problem to that of place-dependent change detection. While we use the method with VT&R, it
can be generalized to suit arbitrary path-following applications.
Visual Teach and Repeat 3 (VT&R3), a generalization of stereo VT&R, achieves
long-term autonomous path-following using topometric mapping and localization
from a single rich sensor stream. In this paper, we improve the capabilities of
a LiDAR implementation of VT&R3 to reliably detect and avoid obstacles in
changing environments. Our architecture simplifies the obstacle-perception
problem to that of place-dependent change detection. We then extend the
behaviour of generic sample-based motion planners to better suit the
teach-and-repeat problem structure by introducing a new edge-cost metric paired
with a curvilinear planning space. The resulting planner generates naturally
smooth paths that avoid local obstacles while minimizing lateral path deviation
to best exploit prior terrain knowledge. While we use the method with VT&R, it
can be generalized to suit arbitrary path-following applications. Experimental
results from online run-time analysis, unit testing, and qualitative
experiments on a differential drive robot show the promise of the technique for
reliable long-term autonomous operation in complex unstructured environments.
Authors: Jordy Sehn, Yuchen Wu, Timothy D. Barfoot.
The statistical and design considerations that pertain to
dose optimization are discussed. The sample size savings range from 16.6% to 27.3%,
depending on the design and scenario, with a mean savings of 22.1%.
The traditional more-is-better dose selection paradigm, developed based on
cytotoxic chemotherapeutics, is often problematic When applied to the
development of novel molecularly targeted agents (e.g., kinase inhibitors,
monoclonal antibodies, and antibody-drug conjugates). The US Food and Drug
Administration (FDA) initiated Project Optimus to reform the dose optimization
and dose selection paradigm in oncology drug development and call for more
attention to benefit-risk consideration.
We systematically investigated the operating characteristics of the seamless
phase 2-3 design as a strategy for dose optimization, where in stage 1
(corresponding to phase 2) patients are randomized to multiple doses, with or
without a control; and in stage 2 (corresponding to phase 3) the efficacy of
the selected optimal dose is evaluated with a randomized concurrent control or
historical control. Depending on whether the concurrent control is included and
the type of endpoints used in stages 1 and 2, we describe four types of
seamless phase 2-3 dose-optimization designs, which are suitable for different
clinical settings. The statistical and design considerations that pertain to
dose optimization are discussed. Simulation shows that dose optimization phase
2-3 designs are able to control the familywise type I error rates and yield
appropriate statistical power with substantially smaller sample size than the
conventional approach. The sample size savings range from 16.6% to 27.3%,
depending on the design and scenario, with a mean savings of 22.1%. Due to the
interim dose selection, the phase 2-3 dose-optimization design is logistically
and operationally more challenging, and should be carefully planned and
implemented to ensure trial integrity.
Authors: Liyun Jiang, Ying Yuan.
We significantly improve performance using properties of the posterior
in our active learning scheme and for the definition of the GP prior. In
particular we account for the expected dynamical range of the posterior in
different dimensionalities. We test our model against a number of synthetic and
cosmological examples.
We present the GPry algorithm for fast Bayesian inference of general
(non-Gaussian) posteriors with a moderate number of parameters. GPry does not
need any pre-training, special hardware such as GPUs, and is intended as a
drop-in replacement for traditional Monte Carlo methods for Bayesian inference.
Our algorithm is based on generating a Gaussian Process surrogate model of the
log-posterior, aided by a Support Vector Machine classifier that excludes
extreme or non-finite values. An active learning scheme allows us to reduce the
number of required posterior evaluations by two orders of magnitude compared to
traditional Monte Carlo inference. Our algorithm allows for parallel
evaluations of the posterior at optimal locations, further reducing wall-clock
times. We significantly improve performance using properties of the posterior
in our active learning scheme and for the definition of the GP prior. In
particular we account for the expected dynamical range of the posterior in
different dimensionalities. We test our model against a number of synthetic and
cosmological examples. GPry outperforms traditional Monte Carlo methods when
the evaluation time of the likelihood (or the calculation of theoretical
observables) is of the order of seconds; for evaluation times of over a minute
it can perform inference in days that would take months using traditional
methods. GPry is distributed as an open source Python package (pip install
gpry) and can also be found at https://github.com/jonaselgammal/GPry.
Authors: Jonas El Gammal, Nils Schöneberg, Jesús Torrado, Christian Fidler.
We consider the fundamental scheduling problem of minimizing the sum of
weighted completion times on a single machine in the non-clairvoyant setting. However, to the best of our knowledge, this concept has never been considered
for the total completion time objective in the non-clairvoyant model. This implies
a performance guarantee of $(1+3\sqrt{3})\approx 6.197$ for the deterministic
algorithm and of $\approx 3.032$ for the randomized version.
We consider the fundamental scheduling problem of minimizing the sum of
weighted completion times on a single machine in the non-clairvoyant setting.
While no non-preemptive algorithm is constant competitive, Motwani, Phillips,
and Torng (SODA '93) proved that the simple preemptive round robin procedure is
$2$-competitive and that no better competitive ratio is possible, initiating a
long line of research focused on preemptive algorithms for generalized variants
of the problem. As an alternative model, Shmoys, Wein, and Williamson (FOCS
'91) introduced kill-and-restart schedules, where running jobs may be killed
and restarted from scratch later, and analyzed then for the makespan objective.
However, to the best of our knowledge, this concept has never been considered
for the total completion time objective in the non-clairvoyant model.
We contribute to both models: First we give for any $b > 1$ a tight analysis
for the natural $b$-scaling kill-and-restart strategy for scheduling jobs
without release dates, as well as for a randomized variant of it. This implies
a performance guarantee of $(1+3\sqrt{3})\approx 6.197$ for the deterministic
algorithm and of $\approx 3.032$ for the randomized version. Second, we show
that the preemptive Weighted Shortest Elapsed Time First (WSETF) rule is
$2$-competitive for jobs released in an online fashion over time, matching the
lower bound by Motwani et al. Using this result as well as the competitiveness
of round robin for multiple machines, we prove performance guarantees of
adaptions of the $b$-scaling algorithm to online release dates and unweighted
jobs on identical parallel machines.
Authors: Sven Jäger, Guillaume Sagnol, Daniel Schmidt genannt Waldschmidt, Philipp Warode.
Frozen pretrained models have become a viable alternative to the
pretraining-then-finetuning paradigm for transfer learning. With this work, we hope to
bring greater attention to this promising path of freezing pretrained image
models.
Frozen pretrained models have become a viable alternative to the
pretraining-then-finetuning paradigm for transfer learning. However, with
frozen models there are relatively few parameters available for adapting to
downstream tasks, which is problematic in computer vision where tasks vary
significantly in input/output format and the type of information that is of
value. In this paper, we present a study of frozen pretrained models when
applied to diverse and representative computer vision tasks, including object
detection, semantic segmentation and video action recognition. From this
empirical analysis, our work answers the questions of what pretraining task
fits best with this frozen setting, how to make the frozen setting more
flexible to various downstream tasks, and the effect of larger model sizes. We
additionally examine the upper bound of performance using a giant frozen
pretrained model with 3 billion parameters (SwinV2-G) and find that it reaches
competitive performance on a varied set of major benchmarks with only one
shared frozen base network: 60.0 box mAP and 52.2 mask mAP on COCO object
detection test-dev, 57.6 val mIoU on ADE20K semantic segmentation, and 81.7
top-1 accuracy on Kinetics-400 action recognition. With this work, we hope to
bring greater attention to this promising path of freezing pretrained image
models.
Authors: Yutong Lin, Ze Liu, Zheng Zhang, Han Hu, Nanning Zheng, Stephen Lin, Yue Cao.
We evaluate the plausibility of these embeddings across different models in
predicting target entities. We also evaluate the meaningfulness of knowledge
proximity to explain the domain expansion profiles of inventors and assignees.
Knowledge proximity refers to the strength of association between any two
entities in a structural form that embodies certain aspects of a knowledge
base. In this work, we operationalize knowledge proximity within the context of
the US Patent Database (knowledge base) using a knowledge graph (structural
form) named PatNet built using patent metadata, including citations, inventors,
assignees, and domain classifications. Using several graph embedding models
(e.g., TransE, RESCAL), we obtain the embeddings of entities and relations that
constitute PatNet. The cosine similarity between the corresponding (or
transformed) embeddings entities denotes the knowledge proximity between these.
We evaluate the plausibility of these embeddings across different models in
predicting target entities. We also evaluate the meaningfulness of knowledge
proximity to explain the domain expansion profiles of inventors and assignees.
We then apply the embeddings of the best-preferred model to associate
homogeneous (e.g., patent-patent) and heterogeneous (e.g., inventor-assignee)
pairs of entities.
Authors: Guangtong Li, L Siddharth, Jianxi Luo.
It resembles a phase transition in that the scalar configuration only appears when a certain quantity that characterizes the compact object, e.g., its compactness or spin, is beyond a threshold.
Scalarization is a mechanism that endows strongly self-gravitating bodies,
such as neutron stars and black holes, with a scalar field configuration. It
resembles a phase transition in that the scalar configuration only appears when
a certain quantity that characterizes the compact object, e.g., its compactness
or spin, is beyond a threshold. We provide a critical and comprehensive review
of scalarization, including the mechanism itself, theories that exhibit it, its
manifestation in neutron stars, black holes, and their binaries, potential
extension to other fields, and a thorough discussion of future perspectives.
Authors: Daniela D. Doneva, Fethi M. Ramazanoğlu, Hector O. Silva, Thomas P. Sotiriou, Stoytcho S. Yazadjiev.
More recently, two new directions for studying
heavy impurity with FDA have been developed. One is to extend FDA to a strongly
correlated background superfluid background, a Bardeen-Cooper-Schrieffer (BCS)
superfluid. Multidimensional Ramsey spectroscopy allows us to investigate correlations
between spectral peaks of an impurity-medium system that is not accessible in
the conventional one-dimensional spectrum.
In this brief review, we report some new development in the functional
determinant approach (FDA), an exact numerical method, in the studies of a
heavy quantum impurity immersed in Fermi gases and manipulated with
radio-frequency pulses. FDA has been successfully applied to investigate the
universal dynamical responses of a heavy impurity in an ultracold ideal Fermi
gas in both the time and frequency domain, which allows the exploration of the
renowned Anderson's orthogonality catastrophe (OC). In such a system, OC is
induced by the multiple particle-hole excitations of the Fermi sea, which is
beyond a simple perturbation picture and manifests itself as the absence of
quasiparticles named polarons. More recently, two new directions for studying
heavy impurity with FDA have been developed. One is to extend FDA to a strongly
correlated background superfluid background, a Bardeen-Cooper-Schrieffer (BCS)
superfluid. In this system, Anderson's orthogonality catastrophe is prohibited
due to the suppression of multiple particle-hole excitations by the superfluid
gap, which leads to the existence of genuine polaron. The other direction is to
generalize the FDA to the case of multiple RF pulses scheme, which extends the
well-established 1D Ramsey spectroscopy in ultracold atoms into
multidimensional, in the same spirit as the well-known multidimensional nuclear
magnetic resonance and optical multidimensional coherent spectroscopy.
Multidimensional Ramsey spectroscopy allows us to investigate correlations
between spectral peaks of an impurity-medium system that is not accessible in
the conventional one-dimensional spectrum.
Authors: Jia Wang.
Nevertheless, in many real-world applications, e.g., magnetohydrodynamics, plasma physics, superconductors, etc. dynamical gauge fields and Coulomb interactions are fundamental. We numerically study the spectrum of the lowest quasi-normal modes and successfully compare the obtained results to magnetohydrodynamics theory in $2+1$ dimensions.
Within the framework of holography, the Einstein-Maxwell action with
Dirichlet boundary conditions corresponds to a dual conformal field theory in
presence of an external gauge field. Nevertheless, in many real-world
applications, e.g., magnetohydrodynamics, plasma physics, superconductors, etc.
dynamical gauge fields and Coulomb interactions are fundamental. In this work,
we consider bottom-up holographic models at finite magnetic field and (free)
charge density in presence of dynamical boundary gauge fields which are
introduced using mixed boundary conditions. We numerically study the spectrum
of the lowest quasi-normal modes and successfully compare the obtained results
to magnetohydrodynamics theory in $2+1$ dimensions. Surprisingly, as far as the
electromagnetic coupling is small enough, we find perfect agreement even in the
large magnetic field limit. Our results prove that a holographic description of
magnetohydrodynamics does not necessarily need higher-form bulk fields but can
be consistently derived using mixed boundary conditions for standard gauge
fields.
Authors: Yongjun Ahn, Matteo Baggioli, Kyoung-Bum Huh, Hyun-Sik Jeong, Keun-Young Kim, Ya-Wen Sun.
We conclude by providing numerical results
comparing our methods to the state of the art.
Inspired by regularization techniques in statistics and machine learning, we
study complementary composite minimization in the stochastic setting. This
problem corresponds to the minimization of the sum of a (weakly) smooth
function endowed with a stochastic first-order oracle, and a structured
uniformly convex (possibly nonsmooth and non-Lipschitz) regularization term.
Despite intensive work on closely related settings, prior to our work no
complexity bounds for this problem were known. We close this gap by providing
novel excess risk bounds, both in expectation and with high probability. Our
algorithms are nearly optimal, which we prove via novel lower complexity bounds
for this class of problems. We conclude by providing numerical results
comparing our methods to the state of the art.
Authors: Alexandre d'Aspremont, Cristóbal Guzmán, Clément Lezane.
We evaluate our proposed approach on the benchmark dataset IEMOCAP, and demonstrate high performance surpassing that in the literature. The code to reproduce the results is available at github.com/skakouros/s3prl_attentive_correlation.
When recognizing emotions from speech, we encounter two common problems: how
to optimally capture emotion-relevant information from the speech signal and
how to best quantify or categorize the noisy subjective emotion labels.
Self-supervised pre-trained representations can robustly capture information
from speech enabling state-of-the-art results in many downstream tasks
including emotion recognition. However, better ways of aggregating the
information across time need to be considered as the relevant emotion
information is likely to appear piecewise and not uniformly across the signal.
For the labels, we need to take into account that there is a substantial degree
of noise that comes from the subjective human annotations. In this paper, we
propose a novel approach to attentive pooling based on correlations between the
representations' coefficients combined with label smoothing, a method aiming to
reduce the confidence of the classifier on the training labels. We evaluate our
proposed approach on the benchmark dataset IEMOCAP, and demonstrate high
performance surpassing that in the literature. The code to reproduce the
results is available at github.com/skakouros/s3prl_attentive_correlation.
Authors: Sofoklis Kakouros, Themos Stafylakis, Ladislav Mosner, Lukas Burget.
A continuous bundle of $C^*$-algebras provides a rigorous framework to study
the thermodynamic limit of quantum theories. a mathematical formalization in which convergence of algebraic quantum
states to probability measures on phase space (typically a Poisson or
symplectic manifold) is studied. We additionally show that the ensuing limit
corresponds to the unique probability measure satisfying the so-called
classical or static KMS- condition.
A continuous bundle of $C^*$-algebras provides a rigorous framework to study
the thermodynamic limit of quantum theories. If the bundle admits the
additional structure of a strict deformation quantization (in the sense of
Rieffel) one is allowed to study the classical limit of the quantum system,
i.e. a mathematical formalization in which convergence of algebraic quantum
states to probability measures on phase space (typically a Poisson or
symplectic manifold) is studied. In this manner we first prove the existence of
the classical limit of Gibbs states illustrated with a class of Schr\"{o}dinger
operators in the regime where Planck's constant $\hbar$ appearing in front of
the Laplacian approaches zero. We additionally show that the ensuing limit
corresponds to the unique probability measure satisfying the so-called
classical or static KMS- condition. Subsequently, a similar study is conducted
for the free energy in the classical limit of mean-field quantum spin systems
in the regime of large particles, and the existence of the classical limit of
the relevant Gibbs states is discussed.
Authors: Christiaan J. F. van de Ven.
Context: Smart TVs have become one of the most popular television types. Many app developers and service providers have designed TV versions for their smartphone applications. The relationship between phone and TV has not been the subject of research works. Method: We gather a large-scale phone/TV app pairs from Google Play Store. ), code (e.g., components, methods, user interactions, etc. ), security and privacy (e.g., reports of AndroBugs and FlowDroid).
Context: Smart TVs have become one of the most popular television types. Many
app developers and service providers have designed TV versions for their
smartphone applications. Despite the extensive studies on mobile app analysis,
its TV equivalents receive far too little attention. The relationship between
phone and TV has not been the subject of research works. Objective: In this
paper, we aim to characterize the relationship between smartphone and smart TV
apps. To fill this gap, we conduct a comparative study on smartphone and smart
TV apps in this work, which is the starting and fundamental step to uncover the
domain-specific challenges. Method: We gather a large-scale phone/TV app pairs
from Google Play Store. We then analyzed the app pairs quantitatively and
qualitatively from a variety of perspectives, including non-code (e.g.,
metadata, resources, permissions, etc.), code (e.g., components, methods, user
interactions, etc.), security and privacy (e.g., reports of AndroBugs and
FlowDroid). Results: Our experimental results indicate that (1) the code of the
smartphone and TV apps can be released in the same app package or in separate
app packages with the same package name; (2) 43% of resource files and 50% of
code methods are reused between phone/TV app pairs; (3) TV and phone versions
of the same app often encounter different kinds of security vulnerabilities;
and (4) TV apps encounter fewer user interactions than their phone versions,
but the type of user interaction events, surprisingly, are similar between
phone/TV apps. Conclution: Our findings are valuable for developers and
academics in comprehending the TV app ecosystem by providing additional insight
into the migration of phone apps to TVs and the design mechanism of analysis
tools for TV apps.
Authors: Yonghui Liu, Xiao Chen, Yue Liu, Pingfan Kong, Tegawendé F. Bissyande, Jacques Klein, Xiaoyu Sun, Chunyang Chen, John Grundy.
Streaming models are an essential component of real-time speech enhancement
tools. We demonstrate that the proposed technique leads to stable
improvement across different architectures and training scenarios.
Streaming models are an essential component of real-time speech enhancement
tools. The streaming regime constrains speech enhancement models to use only a
tiny context of future information, thus, the low-latency streaming setup is
generally assumed to be challenging and has a significant negative effect on
the model quality. However, due to the sequential nature of streaming
generation, it provides a natural possibility for autoregression, i.e., using
previous predictions when making current ones. In this paper, we present a
simple, yet effective trick for training of autoregressive low-latency speech
enhancement models. We demonstrate that the proposed technique leads to stable
improvement across different architectures and training scenarios.
Authors: Pavel Andreev, Nicholas Babaev, Azat Saginbaev, Ivan Shchekotov.
Magneto-optic Kerr effect can probe the process of magnetization reversal in ferromagnetic thin films and thus be used as an alternative to magnetometry. Kerr effect is wavelength-dependent and the Kerr rotation can reverse sign, vanishing at particular wavelengths. We investigate epitaxial heterostructures of ferromagnetic manganite, La$_{0.7}$Sr$_{0.3}$Mn$_{0.9}$Ru$_{0.1}$O$_3$, by polar Kerr effect and magnetometry. The manganite layers are separated by or interfaced with a layer of nickelate, NdNiO$_3$.
Magneto-optic Kerr effect can probe the process of magnetization reversal in
ferromagnetic thin films and thus be used as an alternative to magnetometry.
Kerr effect is wavelength-dependent and the Kerr rotation can reverse sign,
vanishing at particular wavelengths. We investigate epitaxial heterostructures
of ferromagnetic manganite, La$_{0.7}$Sr$_{0.3}$Mn$_{0.9}$Ru$_{0.1}$O$_3$, by
polar Kerr effect and magnetometry. The manganite layers are separated by or
interfaced with a layer of nickelate, NdNiO$_3$. Kerr rotation hysteresis loops
of trilayers, with two manganite layers of different thickness separated by a
nickelate layer, have intriguing humplike features, when measured with light of
400 nm wavelength. By investigating additional reference samples we disentangle
the contributions of the individual layers to the loops: we show that the humps
originate from the opposite sense of the Kerr rotation of the two different
ferromagnetic layers, combined with the additive behavior of the Kerr signal.
Authors: Jörg Schöpf, Paul H. M. van Loosdrecht, Ionela Lindfors-Vrejoiu.