Backward Prediction Research Articles

The Effects of Adverse Life Events on Brain Development in the ABCD Study®: A Propensity-weighted Analysis.

Longitudinal studies of the effects of adversity on human brain development are complicated by the association of stressful events with confounding variables. To counter this bias, we apply a propensity-weighted analysis of the first two years of The Adolescent Brain Cognitive DevelopmentSM (ABCD) Study® data, employing a machine learning analysis weighted by individuals' propensity to experience adversity. Data included 338 resting-state functional connections from 7190 youth (46% female), divided into a training group (80%) and an independent testing group (20%). Propensity scores were computed using 390 variables to balance across two-year adverse life event exposures. Using elastic net regularization with and without inverse propensity weighting, we developed linear models in which changes in functional connectivity of brain connections during the two-year period served as predictors of the number of adverse events experienced during that same period. Haufe's method was applied to forward-transform the backward prediction models. We also tested whether brain changes associated with adverse events correlated with concomitant changes in internalizing or externalizing behaviors or to academic achievement. In the propensity-weighted analysis, brain development significantly predicted the number of adverse events experienced during that period in both the training group (ρ=0.14, p<0.001) and the independent testing group (ρ=0.10, p<0.001). The predictor indicated a general pattern of decreased functional connectivity between large-scale networks and subcortical brain regions, particularly for cingulo-opercular and sensorimotor networks. These network-to-subcortical functional connectivity decreases inversely associated with the development of internalizing symptoms, suggesting adverse events promoted adaptive brain changes that may buffer against stress-related psychopathology. However, these same functional connections were also associated with poorer grades at the two-year follow-up. Although cortical-subcortical brain developmental responses to adversity potentially shield against stress-induced mood and anxiety disorders, they may be detrimental to other domains such as academic success.

Humans adaptively deploy forward and backward prediction.

The formation of predictions is essential to our ability to build models of the world and use them for intelligent decision-making. Here we challenge the dominant assumption that humans form only forward predictions, which specify what future events are likely to follow a given present event. We demonstrate that in some environments, it is more efficient to use backward prediction, which specifies what present events are likely to precede a given future event. This is particularly the case in diverging environments, where possible future events outnumber possible present events. Correspondingly, in six preregistered experiments (n = 1,299) involving both simple decision-making and more challenging planning tasks, we find that humans engage in backward prediction in divergent environments and use forward prediction in convergent environments. We thus establish that humans adaptively deploy forward and backward prediction in the service of efficient decision-making.

Auxiliary two-filter particle smoothing for one generalized hidden Markov model

This paper develops two-filter particle smoothing (TFPS) algorithms for the nonlinear fixed-interval smoothing problem of one generalized hidden Markov model (GHMM), where the current observation depends not only on the current state, but also on one-step previous state. Firstly, by Bayesian approach, the two-filter smoothing (TFS) formula for GHMM is established to calculate smoothing densities. In this TFS formula, the backward information prediction density is generally not a density of the state. This results in a difficulty that the normal sequential Monte Carlo (SMC) sampling technique cannot be directly applied to design corresponding TFPS algorithms based on the TFS formula. To solve this difficulty, a generalized TFS formula for GHMM is then proposed by introducing a sequence of artificial densities. By combining this generalized TFS formula, SMC, and the auxiliary variable sampling technique, a basic auxiliary TFPS (ATFPS) algorithm with quadratic computational complexity is proposed, and a simplified ATFPS algorithm with linear computational complexity is further developed. Finally, the effectiveness and superiority of the two proposed ATFPS algorithms for GHMM are verified via simulation examples and real experimental data.

Bi-directional prediction of hydrothermal carbonization characteristics of agroforestry and livestock wastes with variable components: Graph learning model-aided waste recycle

Agricultural resource management is essential in present-day society. The hydrothermal carbonization (HTC) is a promising technology for recovering valuable resources from wastes. However, there is a paucity of efficient approaches to predict and comprehend the agricultural waste HTC process. Machine learning modeling approaches have emerged as an important tool for simulating thermochemical reaction processes such as HTC. However, previous studies all limit to predicting products properties of single-component HTC, which are contrary to the complex composition of raw materials, and demonstrate poor generalization performance. Herein, the homogeneity graph data is employed to represent the HTC data with variable components. Then, based on graph encoder-decoder transformer method, a bi-directional prediction model is established to model the HTC process of agroforestry and livestock wastes. In the case of the forward prediction, the model can predict the composition and yield of hydrochar according to the given HTC conditions. The comparisons with the experimental data other than the training and validation datasets show that the prediction accuracy (R2) of ash, carbon and nitrogen contents of the hydrochar are 0.895, 0.908 and 0.840, respectively, which proves the good generalizability of the model. Moreover, owing to the embedding vector, which connects the reactant data with the product data in HTC, the model can also achieve backward prediction of HTC reaction conditions such as raw material composition and reaction temperature. Furthermore, by dimensionality reduction analysis of the embedding vector in the model, the similarity of any two HTC reactions can be quantified. The established HTC-graph neural network (GNN) model shows both high accuracy and good generalization ability in bi-directional prediction of HTC process. Based on the effectiveness of the proposed model and the given evaluation indexes, the corresponding HTC parameters can be determined, which greatly benefits to the resources saving and energy consumption reduction.

Distributed Newton method for time-varying convex optimization with backward Euler prediction

&lt;p&gt;We investigated the challenge of unconstrained distributed optimization with a time-varying objective function, employing a prediction-correction approach. Our method introduced a backward Euler prediction step that used the differential information from consecutive moments to forecast the trajectory's future direction. This predicted value was then refined through an iterative correction process. Our analysis and experimental results demonstrated that this approach effectively addresses the optimization problem without requiring the computation of the Hessian matrix's inverse.&lt;/p&gt;

Two-Direction Prediction Method of Drilling Fluid Based on OS-ELM for Water Well Drilling

In this study, a drilling fluid prediction method based on an online sequential extreme learning machine (OS-ELM) is proposed, which is prepared for water well drilling on the muddy clay formation of Tarim Basin, Qinghai Province. First, we investigated the mechanism linking mix ratio to fluid performance, allowing us to employ an OS-ELM algorithm derived from the extreme learning machine. Particularly, the proposed prediction method is bidirectional to identify an appropriate slurry formulation. The forward prediction model is established to predict the fluid performance, where the mud additive contents are inputs, and the drilling fluid properties parameters are outputs. Correspondingly, the backward prediction model is established to modify the slurry formula, where differences in the drilling fluid properties are inputs and percentages of slurry additives amount are output. The simulation results show that the two-direction OS-ELM prediction model can better predict the drilling fluid properties in water well drilling.

Multisource fusion of exogenous inputs based NARXs neural network for vehicle speed prediction between urban road intersections

The economy and safety of passages through the urban road intersection environment is an important research topic in the field of intelligent transportation systems, but vehicle speed prediction as its subtopic is still under-researched, and its prediction accuracy is unsatisfactory. Therefore, a model for vehicle speed prediction based on the nonlinear autoregressive model with multisource exogenous inputs (NARXs) neural network is proposed. The model combines the human-vehicle-road model with the NARXs neural network to perform speed prediction between urban road intersections. First, multisource features, including the variables of driving behavior characteristics, vehicle responses, and road conditions, are extracted to construct the human-vehicle-road model. Then, the model is introduced into the NARXs neural network. Finally, the advantages of the proposed model are verified from two perspectives by evaluation indices such as mean absolute error ( MAE), mean absolute percentage error ( MAPE), root mean square error ( RMSE), Theil index ( Theil ic), and goodness-of-fit ( R2) compared with several other models. On the one hand, the analysis results show that the proposed model has higher prediction accuracy than the other comparative models for different prediction durations and has the best performance in 30 s duration backward prediction. On the other hand, the curves of each evaluation index of the proposed model are horizontal, which indicates that the prediction performance of the model hardly varies with the length of the training dataset. These positive results demonstrate the higher accuracy and outstanding characteristics of the proposed model in the subject of vehicle speed prediction.

Fourier spectral resolution enhancement algorithm based on linear prediction

Fourier transform infrared spectroscopy (FTIR) has been widely used for real-time monitoring of ambient gases, but for quantitative analysis of certain gases, higher spectral resolution is required to quantify more accurately. In this paper, a Fourier spectral resolution enhancement algorithm based on linear prediction is proposed. Firstly, the linear prediction model for interference signals is studied, utilizing the forward and backward linear prediction total least squares method for parameter estimation, minimum error criterion for determining the autoregressive (AR) model order, and sliding window technique for multi-step prediction. Then, the impact of the signal-to-noise ratio, enhancement factor, and initial resolution of the FTIR spectrum on the accuracy of the algorithm was studied through simulations. Finally, the algorithm is applied to spectral resolution enhancement and quantitative analysis of ammonia (NH3). The experimental results show that when the spectral resolution of NH3 is enhanced from 4 cm−1 to 2 cm−1 and 1 cm−1, the spectral error of the characteristic band is 0.16% and 0.26%, respectively. Additionally, the accuracy of concentration inversion is improved by 2.13% and 4.06%, respectively. The relative error of concentration inversion is 0.69% and 2.15%, respectively, when compared to the measured spectrum at corresponding resolutions. The algorithm proves effective in enhancing spectral resolution and improving the accuracy of quantitative analysis.

On the efficiency of growth forecasts for Germany: an application of forward and backward predictor variable selection

ABSTRACT I use a forward and backward predictor selection to study the efficiency of growth forecasts for Germany published by three German economic research institutes during the sample period 1970–2018. The two predictor selection techniques make it straightforward to study forecast efficiency even when the number of predictor variables that a researcher uses to proxy the information set of a forecaster is large. Based on my empirical results, I reject forecast efficiency in the majority of cases.

Inverted pyramid frame forward and backward prediction for distorted video by water waves.

There has been much research on how to restore a single image from distorted video. Random water surface variation, an inability to model the surface, and multiple factors in the imaging processing leading to different geometric distortions in each frame are among the challenges. This paper proposes an inverted pyramid structure based on the cross optical flow registration approach and a multi-scale weight fusion method based on wavelet decomposition. The inverted pyramid based on the registration method is used to estimate the original pixel positions. A multi-scale image fusion method is applied to fuse the two inputs processed by optical flow and backward mapping, and two iterations are proposed to improve the accuracy and stability of the output video. The method is tested on several reference distorted videos and our videos, which were obtained through our experimental equipment. The obtained results exhibit significant improvements over other reference methods. The corrected videos obtained with our approach have a higher degree of sharpness, and the time required to restore the videos is significantly reduced.

Algorithms for Sparse Multichannel Blind Deconvolution

In this paper we present two algorithms for sparse multichannel blind deconvolution. The first algorithm is based on a cascade of a forward and a backward prediction error filter (C-PEF). The second consists in an alternating minimization algorithm for estimating both the reflectivity series and the seismic wavelet (AM-SMBD). We also compare the algorithms with other state-of-the-art sparse blind deconvolution algorithms. Simulation results with synthetic data for different SNR levels showed that the AM-SMBD was outperformed (in terms of the Pearson Correlation Coefficient and the Gini Correlation Coefficient) other estimation methods, like the reduced SMBD, the TS-sparseTLS (Toeplitz-structured sparse total least square) and the SMBD-SPG (Sparse multichannel blind deconvolution via spectral projected-gradient). For the same data, the C-PEF was able to provide better results (in terms of the Gini Correlation Coefficient, visual inspection and frequency gain) when compared with the F-SMBD (Fast sparse multichannel blind deconvolution). In a simulation considering reflectivities with different levels of sparsity, the C-PEF seems to be more robust for less sparse data when compared with AM-SMBD and SMBD-SPG (up to a certain degree of sparsity). Finally, simulations considering a real land acquisition show that both algorithms were able to greatly improve the resolution of the seismic data.

Four-dimensional measurement of root system development using time-series three-dimensional volumetric data analysis by backward prediction.

Root system architecture (RSA) is an essential characteristic for efficient water and nutrient absorption in terrestrial plants; its plasticity enables plants to respond to different soil environments. Better understanding of root plasticity is important in developing stress-tolerant crops. Non-invasive techniques that can measure roots in soils nondestructively, such as X-ray computed tomography (CT), are useful to evaluate RSA plasticity. However, although RSA plasticity can be measured by tracking individual root growth, only a few methods are available for tracking individual roots from time-series three-dimensional (3D) images. We developed a semi-automatic workflow that tracks individual root growth by vectorizing RSA from time-series 3D images via two major steps. The first step involves 3D alignment of the time-series RSA images by iterative closest point registration with point clouds generated by high-intensity particles in potted soils. This alignment ensures that the time-series RSA images overlap. The second step consists of backward prediction of vectorization, which is based on the phenomenon that the root length of the RSA vector at the earlier time point is shorter than that at the last time point. In other words, when CT scanning is performed at time point A and again at time point B for the same pot, the CT data and RSA vectors at time points A and B will almost overlap, but not where the roots have grown. We assumed that given a manually created RSA vector at the last time point of the time series, all RSA vectors except those at the last time point could be automatically predicted by referring to the corresponding RSA images. Using 21 time-series CT volumes of a potted plant of upland rice (Oryza sativa), this workflow revealed that the root elongation speed increased with age. Compared with a workflow that does not use backward prediction, the workflow with backward prediction reduced the manual labor time by 95%. We developed a workflow to efficiently generate time-series RSA vectors from time-series X-ray CT volumes. We named this workflow 'RSAtrace4D' and are confident that it can be applied to the time-series analysis of RSA development and plasticity.

Applying Deep Learning in the Prediction of Chlorophyll-a in the East China Sea

The ocean chlorophyll-a (Chl-a) concentration is an important variable in the marine environment, the abnormal distribution of which is closely related to the hazards of red tides. Thus, the accurate prediction of its concentration in the East China Sea (ECS) is greatly important for preventing water eutrophication and protecting the coastal ecological environment. Processed by two different pre-processing methods, 10-year (2011–2020) satellite-observed chlorophyll-a data and logarithmic data were used as the long short-term memory (LSTM) neural network training datasets in this study. The 2021 data were used for comparison to prediction results. The past 15 days’ data were used to predict the concentration of chlorophyll-a for the five following days. Results showed that the predictions obtained by both pre-processing methods could simulate the seasonal distribution of the Chl-a concentration in the ECS effectively. Moreover, the prediction performance of the model driven by the original values was better in the medium- and low-concentration regions. However, in the high-concentration region, the prediction of extreme concentrations by the two data-driven LSTM models showed underestimation, considering that the prediction performance of the model driven by the original values was better. Results of sensitivity experiments showed that the prediction accuracy of the model decreased considerably when the backward prediction time step increased. In this study, the neural network was driven only by chlorophyll-a, whose concentration in the ECS was forecasted, and the effect of other relevant marine elements on Chl-a was not considered, which is the current weakness of this study.

Canalized gene expression during development mediates caste differentiation in ants

Ant colonies are higher-level organisms consisting of specialized reproductive and non-reproductive individuals that differentiate early in development, similar to germ–soma segregation in bilateral Metazoa. Analogous to diverging cell lines, developmental differentiation of individual ants has often been considered in epigenetic terms but the sets of genes that determine caste phenotypes throughout larval and pupal development remain unknown. Here, we reconstruct the individual developmental trajectories of two ant species, Monomorium pharaonis and Acromyrmex echinatior, after obtaining >1,400 whole-genome transcriptomes. Using a new backward prediction algorithm, we show that caste phenotypes can be accurately predicted by genome-wide transcriptome profiling. We find that caste differentiation is increasingly canalized from early development onwards, particularly in germline individuals (gynes/queens) and that the juvenile hormone signalling pathway plays a key role in this process by regulating body mass divergence between castes. We quantified gene-specific canalization levels and found that canalized genes with gyne/queen-biased expression were enriched for ovary and wing functions while canalized genes with worker-biased expression were enriched in brain and behavioural functions. Suppression in gyne larvae of Freja, a highly canalized gyne-biased ovary gene, disturbed pupal development by inducing non-adaptive intermediate phenotypes between gynes and workers. Our results are consistent with natural selection actively maintaining canalized caste phenotypes while securing robustness in the life cycle ontogeny of ant colonies.

Adaptive Lower Limb Pattern Recognition for Multi-Day Control.

Pattern recognition in EMG-based control systems suffer from increase in error rate over time, which could lead to unwanted behavior. This so-called concept drift in myoelectric control systems could be caused by fatigue, sensor replacement and varying skin conditions. To circumvent concept drift, adaptation strategies could be used to retrain a pattern recognition system, which could lead to comparable error rates over multiple days. In this study, we investigated the error rate development over one week and compared three adaptation strategies to reduce the error rate increase. The three adaptation strategies were based on entropy, on backward prediction and a combination of backward prediction and entropy. Ten able-bodied subjects were measured on four measurement days while performing gait-related activities. During the measurement electromyography and kinematics were recorded. The three adaptation strategies were implemented and compared against the baseline error rate and against adaptation using the ground truth labels. It can be concluded that without adaptation the baseline error rate increases significantly from day 1 to 2, but plateaus on day 2, 3 and 7. Of the three tested adaptation strategies, entropy based adaptation showed the smallest increase in error rate over time. It can be concluded that entropy based adaptation is simple to implement and can be considered a feasible adaptation strategy for lower limb pattern recognition.

Autoregressive Reconstruction of Total Water Storage within GRACE and GRACE Follow-On Gap Period

For 15 years, the Gravity Recovery and Climate Experiment (GRACE) mission have monitored total water storage (TWS) changes. The GRACE mission ended in October 2017, and 11 months later, the GRACE Follow-On (GRACE-FO) mission was launched in May 2018. Bridging the gap between both missions is essential to obtain continuous mass changes. To fill the gap, we propose a new approach based on a remove–restore technique combined with an autoregressive (AR) prediction. We first make use of the Global Land Data Assimilation System (GLDAS) hydrological model to remove climatology from GRACE/GRACE-FO data. Since the GLDAS mis-models real TWS changes for many regions around the world, we further use least-squares estimation (LSE) to remove remaining residual trends and annual and semi-annual oscillations. The missing 11 months of TWS values are then predicted forward and backward with an AR model. For the forward approach, we use the GRACE TWS values before the gap; for the backward approach, we use the GRACE-FO TWS values after the gap. The efficiency of forward–backward AR prediction is examined for the artificial gap of 11 months that we create in the GRACE TWS changes for the July 2008 to May 2009 period. We obtain average differences between predicted and observed GRACE values of at maximum 5 cm for 80% of areas, with the extreme values observed for the Amazon, Alaska, and South and Northern Asia. We demonstrate that forward–backward AR prediction is better than the standalone GLDAS hydrological model for more than 75% of continental areas. For the natural gap (July 2017–May 2018), the misclosures in backward–forward prediction estimated between forward- and backward-predicted values are equal to 10 cm. This represents an amount of 10–20% of the total TWS signal for 60% of areas. The regional analysis shows that the presented method is able to capture the occurrence of droughts or floods, but does not reflect their magnitudes. Results indicate that the presented remove–restore technique combined with AR prediction can be utilized to reliably predict TWS changes for regional analysis, but the removed climatology must be properly matched to the selected region.

Research of Quantative Trading Strategy Based on LSTM

Making strategies to maximize returns has always been the biggest problem almost every investor faces in the process of investment. In this paper, we will establish a model to analyze the internal characteristics of gold and bitcoin price data for further price prediction, which is of great help in formulation of trading strategy for the future. In Section 2 we hope to analyze the price trend of gold and bitcoin. At the beginning we establish a model based on ARIMA and get a relatively good prediction result. However, considering the errors may occur when the backward prediction unit gets long in this model, we try to use LSTM algorithm to better extract the nonlinear characteristics of the two time series of price. Through LSTM algorithm we get a much better result, and thus we choose LSTM as our final prediction model after comprehensive comparison. In Section 3, we use the predicted results to make trading strategies. We first divide investors into reckless and prudent types and establish models under different assumptions respectively. Then by applying greedy algorithm and establishing risk quantification model we approach the global optimal solution respectively. In the end, after testing the model and confirming its excellent performance, we summarize the advantages and shortcomings about this model comprehensively. Moreover, much more conclusions are drawn to analyze the possible model update in future.Price Prediction

Action-dependent bidirectional contrastive predictive coding for neural belief representations

The key to solving the complex, partially observable Markov decision process (POMDP) with high-dimensional observations lies in explicit belief representations. However, the existing methods generally adopted the black-box model for shaping beliefs, which is inefficient and lacks interpretability. Due to this reason, the action-dependent bidirectional contrastive predictive coding (BCPC|Action) is proposed in this paper, in which the observation features are extracted efficiently through self-supervised contrastive learning. Owing to the bottleneck belief constraints in the bidirectional model, the upper bound of prediction errors is effectively reduced. Besides, the forward prediction is optimized by the guidance of an easier trainable backward prediction; thus, the bidirectional match regularization (BMR) could be derived for stabilizing the training process. More importantly, the interpretability of the learned belief representation is thoroughly explored based on the gradient truncation. Simulation results verify the effectiveness of the presented method; apart from achieving highly accurate belief tracking, the state uncertainties could be characterized reasonably, which provides a guarantee for solving the POMDP optimal policy for downstream tasks.

COLF‐GAN: Learning to axial super‐resolve focal stacks

In order to generate a denser focal stack, a cooperative generative adversarial network is proposed to learn the refocusing ability from light field imaging. To keep the axial continuity, the proposed framework is designed to learn features of a focal stack in both axial directions. Different from the classic generative adversarial network, our generative module consists of a forward prediction sub-network and a backward prediction sub-network, taking the forward-and-backward focal stacks as their inputs, respectively. The bi-directional predictions are then fused by a weighting process, which is guided by an adversarial module. The proposed network is trained on light field focal stacks conducted via digital refocusing. Without loss of the refocus continuity, one can axial super-resolve a focal stack by using the trained model. The effectiveness of the proposed algorithm on different types of focal stacks produced by both light fields and traditional camera shootings is validated. The experimental results indicate that the refocus variation of a focal stack can be well learned and predicted without a complete light field. Therefore, the proposed algorithm outperforms the traditional digital refocusing in terms of run-time.

For-backward LSTM-based missing data reconstruction for time-series Landsat images

ABSTRACT Reconstructing the missing data for cloud/shadow-covered optical satellite images has great significance for enhancing the data availability and multi-temporal analysis. In this study, we proposed a deep-learning-based method for cloud/shadow-covered missing data reconstruction for time-series Landsat images. Central to this method is the combined use of autoencoder, long-short-term memory (AE-LSTM)-based similar pixel clustering and for backward LSTM-based time-series prediction. First, manually delineated cloud/shadow-covered masks were overlaid onto multi-temporal satellite images to produce pixel-wise time-series data with masking values. Second, these pixel-wise time series were clustered by an AE-LSTM-based unsupervised method into multiple clusters, for searching similar pixels. Third, for each cluster of target images, a for-backward-LSTM-based model was established to restore missing values in time series data. Finally, reconstructed data were merged with cloud-free (image) regions to produce cloud-free time-series images. The proposed method was applied onto three datasets of multi-temporal Landsat-8 OLI images to restore cloud/shadow-covered images. The reconstruction results, showing an improvement of greater than 10% in normalized mean-square error compared to the state-of-the-art methods, demonstrated the effectiveness of the proposed method in time-series reconstruction for satellite images.