We show how machine learning techniques can be applied for the classification of topological phases in leaky photonic lattices using limited measurement data. We propose an approach based solely on bulk intensity measurements, thus exempt from the need for complicated phase retrieval procedures. In particular, we design a fully connected neural network that accurately determines topological properties from the output intensity distribution in dimerized waveguide arrays with leaky channels, after propagation of a spatially localized initial excitation at a finite distance, in a setting that closely emulates realistic experimental conditions.
相關內容
We present an algorithm for computing melting points by autonomously learning from coexistence simulations in the NPT ensemble. Given the interatomic interaction model, the method makes decisions regarding the number of atoms and temperature at which to conduct simulations, and based on the collected data predicts the melting point along with the uncertainty, which can be systematically improved with more data. We demonstrate how incorporating physical models of the solid-liquid coexistence evolution enhances the algorithm's accuracy and enables optimal decision-making to effectively reduce predictive uncertainty. To validate our approach, we compare the results of 20 melting point calculations from the literature to the results of our calculations, all conducted with same interatomic potentials. Remarkably, we observe significant deviations in about one-third of the cases, underscoring the need for accurate and reliable algorithms for materials property calculations.
We formulate and solve data-driven aerodynamic shape design problems with distributionally robust optimization (DRO) approaches. Building on the findings of the work \cite{gotoh2018robust}, we study the connections between a class of DRO and the Taguchi method in the context of robust design optimization. Our preliminary computational experiments on aerodynamic shape optimization in transonic turbulent flow show promising design results.
In theoretical neuroscience, recent work leverages deep learning tools to explore how some network attributes critically influence its learning dynamics. Notably, initial weight distributions with small (resp. large) variance may yield a rich (resp. lazy) regime, where significant (resp. minor) changes to network states and representation are observed over the course of learning. However, in biology, neural circuit connectivity generally has a low-rank structure and therefore differs markedly from the random initializations generally used for these studies. As such, here we investigate how the structure of the initial weights, in particular their effective rank, influences the network learning regime. Through both empirical and theoretical analyses, we discover that high-rank initializations typically yield smaller network changes indicative of lazier learning, a finding we also confirm with experimentally-driven initial connectivity in recurrent neural networks. Conversely, low-rank initialization biases learning towards richer learning. Importantly, however, as an exception to this rule, we find lazier learning can still occur with a low-rank initialization that aligns with task and data statistics. Our research highlights the pivotal role of initial weight structures in shaping learning regimes, with implications for metabolic costs of plasticity and risks of catastrophic forgetting.
Learning curves are a measure for how the performance of machine learning models improves given a certain volume of training data. Over a wide variety of applications and models it was observed that learning curves follow -- to a large extent -- a power law behavior. This makes the performance of different models for a given task somewhat predictable and opens the opportunity to reduce the training time for practitioners, who are exploring the space of possible models and hyperparameters for the problem at hand. By estimating the learning curve of a model from training on small subsets of data only the best models need to be considered for training on the full dataset. How to choose subset sizes and how often to sample models on these to obtain estimates is however not researched. Given that the goal is to reduce overall training time strategies are needed that sample the performance in a time-efficient way and yet leads to accurate learning curve estimates. In this paper we formulate the framework for these strategies and propose several strategies. Further we evaluate the strategies for simulated learning curves and in experiments with popular datasets and models for image classification tasks.
We investigate a family of approximate multi-step proximal point methods, accelerated by implicit linear discretizations of gradient flow. The resulting methods are multi-step proximal point methods, with similar computational cost in each update as the proximal point method. We explore several optimization methods where applying an approximate multistep proximal points method results in improved convergence behavior. We argue that this is the result of the lowering of truncation error in approximating gradient flow
Recurrent neural networks (RNNs) have yielded promising results for both recognizing objects in challenging conditions and modeling aspects of primate vision. However, the representational dynamics of recurrent computations remain poorly understood, especially in large-scale visual models. Here, we studied such dynamics in RNNs trained for object classification on MiniEcoset, a novel subset of ecoset. We report two main insights. First, upon inference, representations continued to evolve after correct classification, suggesting a lack of the notion of being ``done with classification''. Second, focusing on ``readout zones'' as a way to characterize the activation trajectories, we observe that misclassified representations exhibit activation patterns with lower L2 norm, and are positioned more peripherally in the readout zones. Such arrangements help the misclassified representations move into the correct zones as time progresses. Our findings generalize to networks with lateral and top-down connections, and include both additive and multiplicative interactions with the bottom-up sweep. The results therefore contribute to a general understanding of RNN dynamics in naturalistic tasks. We hope that the analysis framework will aid future investigations of other types of RNNs, including understanding of representational dynamics in primate vision.
Deep Learning and transfer learning models are being used to generate time series forecasts; however, there is scarce evidence about their performance prediction that it is more evident for monthly time series. The purpose of this paper is to compare Deep Learning models with transfer learning and without transfer learning and other traditional methods used for monthly forecasts to answer three questions about the suitability of Deep Learning and Transfer Learning to generate predictions of time series. Time series of M4 and M3 competitions were used for the experiments. The results suggest that deep learning models based on TCN, LSTM, and CNN with transfer learning tend to surpass the performance prediction of other traditional methods. On the other hand, TCN and LSTM, trained directly on the target time series, got similar or better performance than traditional methods for some forecast horizons.
Computational platforms for high-performance scientific applications are becoming more heterogenous, including hardware accelerators such as multiple GPUs. Applications in a wide variety of scientific fields require an efficient and careful management of the computational resources of this type of hardware to obtain the best possible performance. However, there are currently different GPU vendors, architectures and families that can be found in heterogeneous clusters or machines. Programming with the vendor provided languages or frameworks, and optimizing for specific devices, may become cumbersome and compromise portability to other systems. To overcome this problem, several proposals for high-level heterogeneous programming have appeared, trying to reduce the development effort and increase functional and performance portability, specifically when using GPU hardware accelerators. This paper evaluates the SYCL programming model, using the Open SYCL compiler, from two different perspectives: The performance it offers when dealing with single or multiple GPU devices from the same or different vendors, and the development effort required to implement the code. We use as case of study the Finite Time Lyapunov Exponent calculation over two real-world scenarios and compare the performance and the development effort of its Open SYCL-based version against the equivalent versions that use CUDA or HIP. Based on the experimental results, we observe that the use of SYCL does not lead to a remarkable overhead in terms of the GPU kernels execution time. In general terms, the Open SYCL development effort for the host code is lower than that observed with CUDA or HIP. Moreover, the SYCL version can take advantage of both CUDA and AMD GPU devices simultaneously much easier than directly using the vendor-specific programming solutions.
Most state-of-the-art machine learning techniques revolve around the optimisation of loss functions. Defining appropriate loss functions is therefore critical to successfully solving problems in this field. We present a survey of the most commonly used loss functions for a wide range of different applications, divided into classification, regression, ranking, sample generation and energy based modelling. Overall, we introduce 33 different loss functions and we organise them into an intuitive taxonomy. Each loss function is given a theoretical backing and we describe where it is best used. This survey aims to provide a reference of the most essential loss functions for both beginner and advanced machine learning practitioners.
Deep learning constitutes a recent, modern technique for image processing and data analysis, with promising results and large potential. As deep learning has been successfully applied in various domains, it has recently entered also the domain of agriculture. In this paper, we perform a survey of 40 research efforts that employ deep learning techniques, applied to various agricultural and food production challenges. We examine the particular agricultural problems under study, the specific models and frameworks employed, the sources, nature and pre-processing of data used, and the overall performance achieved according to the metrics used at each work under study. Moreover, we study comparisons of deep learning with other existing popular techniques, in respect to differences in classification or regression performance. Our findings indicate that deep learning provides high accuracy, outperforming existing commonly used image processing techniques.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191