We discuss probabilistic neural networks with a fixed internal representation as models for machine understanding. Here understanding is intended as mapping data to an already existing representation which encodes an {\em a priori} organisation of the feature space. We derive the internal representation by requiring that it satisfies the principles of maximal relevance and of maximal ignorance about how different features are combined. We show that, when hidden units are binary variables, these two principles identify a unique model -- the Hierarchical Feature Model (HFM) -- which is fully solvable and provides a natural interpretation in terms of features. We argue that learning machines with this architecture enjoy a number of interesting properties, like the continuity of the representation with respect to changes in parameters and data, the possibility to control the level of compression and the ability to support functions that go beyond generalisation. We explore the behaviour of the model with extensive numerical experiments and argue that models where the internal representation is fixed reproduce a learning modality which is qualitatively different from that of traditional models such as Restricted Boltzmann Machines.
相關內容
Most existing neural network-based approaches for solving stochastic optimal control problems using the associated backward dynamic programming principle rely on the ability to simulate the underlying state variables. However, in some problems, this simulation is infeasible, leading to the discretization of state variable space and the need to train one neural network for each data point. This approach becomes computationally inefficient when dealing with large state variable spaces. In this paper, we consider a class of this type of stochastic optimal control problems and introduce an effective solution employing multitask neural networks. To train our multitask neural network, we introduce a novel scheme that dynamically balances the learning across tasks. Through numerical experiments on real-world derivatives pricing problems, we prove that our method outperforms state-of-the-art approaches.
Tracking ripening tomatoes is time consuming and labor intensive. Artificial intelligence technologies combined with those of computer vision can help users optimize the process of monitoring the ripening status of plants. To this end, we have proposed a tomato ripening monitoring approach based on deep learning in complex scenes. The objective is to detect mature tomatoes and harvest them in a timely manner. The proposed approach is declined in two parts. Firstly, the images of the scene are transmitted to the pre-processing layer. This process allows the detection of areas of interest (area of the image containing tomatoes). Then, these images are used as input to the maturity detection layer. This layer, based on a deep neural network learning algorithm, classifies the tomato thumbnails provided to it in one of the following five categories: green, brittle, pink, pale red, mature red. The experiments are based on images collected from the internet gathered through searches using tomato state across diverse languages including English, German, French, and Spanish. The experimental results of the maturity detection layer on a dataset composed of images of tomatoes taken under the extreme conditions, gave a good classification rate.
Sparsity is a highly desired feature in deep neural networks (DNNs) since it ensures numerical efficiency, improves the interpretability of models (due to the smaller number of relevant features), and robustness. In machine learning approaches based on linear models, it is well known that there exists a connecting path between the sparsest solution in terms of the $\ell^1$ norm,i.e., zero weights and the non-regularized solution, which is called the regularization path. Very recently, there was a first attempt to extend the concept of regularization paths to DNNs by means of treating the empirical loss and sparsity ($\ell^1$ norm) as two conflicting criteria and solving the resulting multiobjective optimization problem. However, due to the non-smoothness of the $\ell^1$ norm and the high number of parameters, this approach is not very efficient from a computational perspective. To overcome this limitation, we present an algorithm that allows for the approximation of the entire Pareto front for the above-mentioned objectives in a very efficient manner. We present numerical examples using both deterministic and stochastic gradients. We furthermore demonstrate that knowledge of the regularization path allows for a well-generalizing network parametrization.
Qualitative probabilistic networks (QPNs) combine the conditional independence assumptions of Bayesian networks with the qualitative properties of positive and negative dependence. They formalise various intuitive properties of positive dependence to allow inferences over a large network of variables. However, we will demonstrate in this paper that, due to an incorrect symmetry property, many inferences obtained in non-binary QPNs are not mathematically true. We will provide examples of such incorrect inferences and briefly discuss possible resolutions.
Recent lay language generation systems have used Transformer models trained on a parallel corpus to increase health information accessibility. However, the applicability of these models is constrained by the limited size and topical breadth of available corpora. We introduce CELLS, the largest (63k pairs) and broadest-ranging (12 journals) parallel corpus for lay language generation. The abstract and the corresponding lay language summary are written by domain experts, assuring the quality of our dataset. Furthermore, qualitative evaluation of expert-authored plain language summaries has revealed background explanation as a key strategy to increase accessibility. Such explanation is challenging for neural models to generate because it goes beyond simplification by adding content absent from the source. We derive two specialized paired corpora from CELLS to address key challenges in lay language generation: generating background explanations and simplifying the original abstract. We adopt retrieval-augmented models as an intuitive fit for the task of background explanation generation, and show improvements in summary quality and simplicity while maintaining factual correctness. Taken together, this work presents the first comprehensive study of background explanation for lay language generation, paving the path for disseminating scientific knowledge to a broader audience. CELLS is publicly available at: //github.com/LinguisticAnomalies/pls_retrieval.
Asymptotic analysis for related inference problems often involves similar steps and proofs. These intermediate results could be shared across problems if each of them is made self-contained and easily identified. However, asymptotic analysis using Taylor expansions is limited for result borrowing because it is a step-to-step procedural approach. This article introduces EEsy, a modular system for estimating finite and infinitely dimensional parameters in related inference problems. It is based on the infinite-dimensional Z-estimation theorem, Donsker and Glivenko-Cantelli preservation theorems, and weight calibration techniques. This article identifies the systematic nature of these tools and consolidates them into one system containing several modules, which can be built, shared, and extended in a modular manner. This change to the structure of method development allows related methods to be developed in parallel and complex problems to be solved collaboratively, expediting the development of new analytical methods. This article considers four related inference problems -- estimating parameters with random sampling, two-phase sampling, auxiliary information incorporation, and model misspecification. We illustrate this modular approach by systematically developing 9 parameter estimators and 18 variance estimators for the four related inference problems regarding semi-parametric additive hazards models. Simulation studies show the obtained asymptotic results for these 27 estimators are valid. In the end, I describe how this system can simplify the use of empirical process theory, a powerful but challenging tool to be adopted by the broad community of methods developers. I discuss challenges and the extension of this system to other inference problems.
We introduce a new model which can be considered as a extended version of the Hawkes process in a discrete sense. This model enables the integration of various residual distributions while preserving the fundamental properties of the original Hawkes process. The rich nature of this model enables a filtered historical simulation which incorporate the properties of original time series more accurately. The process naturally extends to multi-variate models with easy implementations of estimation and simulation. We investigate the effect of flexible residual distribution on estimation of high frequency financial data compared with the Hawkes process.
Remotely sensed data are dominated by mixed Land Use and Land Cover (LULC) types. Spectral unmixing (SU) is a key technique that disentangles mixed pixels into constituent LULC types and their abundance fractions. While existing studies on Deep Learning (DL) for SU typically focus on single time-step hyperspectral (HS) or multispectral (MS) data, our work pioneers SU using MODIS MS time series, addressing missing data with end-to-end DL models. Our approach enhances a Long-Short Term Memory (LSTM)-based model by incorporating geographic, topographic (geo-topographic), and climatic ancillary information. Notably, our method eliminates the need for explicit endmember extraction, instead learning the input-output relationship between mixed spectra and LULC abundances through supervised learning. Experimental results demonstrate that integrating spectral-temporal input data with geo-topographic and climatic information significantly improves the estimation of LULC abundances in mixed pixels. To facilitate this study, we curated a novel labeled dataset for Andalusia (Spain) with monthly MODIS multispectral time series at 460m resolution for 2013. Named Andalusia MultiSpectral MultiTemporal Unmixing (Andalusia-MSMTU), this dataset provides pixel-level annotations of LULC abundances along with ancillary information. The dataset (//zenodo.org/records/7752348) and code (//github.com/jrodriguezortega/MSMTU) are available to the public.
Ordinal pattern dependence has been introduced in order to capture co-monotonic behavior between two time series. This concept has several features one would intuitively demand from a dependence measure. It was believed that ordinal pattern dependence satisfies the axioms which Grothe et al. [8] proclaimed for a multivariate measure of dependence. In the present article we show that this is not true and that there is a mistake in the article Betken et al. [5]. Furthermore we show that ordinal pattern dependence satisfies a slightly modified set of axioms.
Languages have long been described according to their perceived rhythmic attributes. The associated typologies are of interest in psycholinguistics as they partly predict newborns' abilities to discriminate between languages and provide insights into how adult listeners process non-native languages. Despite the relative success of rhythm metrics in supporting the existence of linguistic rhythmic classes, quantitative studies have yet to capture the full complexity of temporal regularities associated with speech rhythm. We argue that deep learning offers a powerful pattern-recognition approach to advance the characterization of the acoustic bases of speech rhythm. To explore this hypothesis, we trained a medium-sized recurrent neural network on a language identification task over a large database of speech recordings in 21 languages. The network had access to the amplitude envelopes and a variable identifying the voiced segments, assuming that this signal would poorly convey phonetic information but preserve prosodic features. The network was able to identify the language of 10-second recordings in 40% of the cases, and the language was in the top-3 guesses in two-thirds of the cases. Visualization methods show that representations built from the network activations are consistent with speech rhythm typologies, although the resulting maps are more complex than two separated clusters between stress and syllable-timed languages. We further analyzed the model by identifying correlations between network activations and known speech rhythm metrics. The findings illustrate the potential of deep learning tools to advance our understanding of speech rhythm through the identification and exploration of linguistically relevant acoustic feature spaces.
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