Ali Karkehabadi

Ali Karkehabadi, Houman Homayoun, Avesta Sasan

Deep Neural Networks are powerful tools for understanding complex patterns and making decisions. However, their black-box nature impedes a complete understanding of their inner workings. Saliency-Guided Training (SGT) methods try to highlight the prominent features in the model's training based on the output to alleviate this problem. These methods use back-propagation and modified gradients to guide the model toward the most relevant features while keeping the impact on the prediction accuracy negligible. SGT makes the model's final result more interpretable by masking input partially. In this way, considering the model's output, we can infer how each segment of the input affects the output. In the particular case of image as the input, masking is applied to the input pixels. However, the masking strategy and number of pixels which we mask, are considered as a hyperparameter. Appropriate setting of masking strategy can directly affect the model's training. In this paper, we focus on this issue and present our contribution. We propose a novel method to determine the optimal number of masked images based on input, accuracy, and model loss during the training. The strategy prevents information loss which leads to better accuracy values. Also, by integrating the model's performance in the strategy formula, we show that our model represents the salient features more meaningful. Our experimental results demonstrate a substantial improvement in both model accuracy and the prominence of saliency, thereby affirming the effectiveness of our proposed solution.

Ali Karkehabadi, Jamshid Hassanpour, Houman Homayoun et al.

Gradient-based saliency methods are widely used to interpret deep neural networks, yet they often produce noisy and unstable explanations that poorly align with semantically meaningful input features. We argue that a fundamental cause of this behavior lies in the geometry of learned representations: correlated feature dimensions diffuse attribution gradients across redundant directions, resulting in blurred and unreliable saliency maps. To address this issue, we identify feature correlation as a structural limitation of gradient-based interpretability and propose SaliencyDecor, a training framework that enforces feature decorrelation to improve attribution fidelity without modifying saliency methods or model architectures by reshaping the feature space toward orthogonality, our approach promotes more concentrated gradient flow and improves the fidelity of saliency-based explanations. SaliencyDecor jointly optimizes classification, prediction consistency under feature masking, and a decorrelation regularizer, requiring no architectural changes or inference-time overhead. Extensive experiments across multiple benchmarks and architectures demonstrate that our method produces substantially sharper and more object-focused saliency maps while simultaneously improving predictive performance, achieving accuracy gains across the datasets. These results establish our method as a principled mechanism for enhancing both interpretability and accuracy, challenging the conventional trade-off between explanation quality and model performance.

Faraz Sasani, Ramin Mousa, Ali Karkehabadi et al.

Stock market forecasting has been a challenging part for many analysts and researchers. Trend analysis, statistical techniques, and movement indicators have traditionally been used to predict stock price movements, but text extraction has emerged as a promising method in recent years. The use of neural networks, especially recurrent neural networks, is abundant in the literature. In most studies, the impact of different users was considered equal or ignored, whereas users can have other effects. In the current study, we will introduce TM-vector and then use this vector to train an IndRNN and ultimately model the market users' behaviour. In the proposed model, TM-vector is simultaneously trained with both the extracted Twitter features and market information. Various factors have been used for the effectiveness of the proposed forecasting approach, including the characteristics of each individual user, their impact on each other, and their impact on the market, to predict market direction more accurately. Dow Jones 30 index has been used in current work. The accuracy obtained for predicting daily stock changes of Apple is based on various models, closed to over 95\% and for the other stocks is significant. Our results indicate the effectiveness of TM-vector in predicting stock market direction.

Rayhaneh Shabani Nia, Ali Karkehabadi

Deep neural networks have achieved strong performance in genomic sequence classification; however, relating their predictions to biologically meaningful sequence patterns remains challenging. In this work, we present AttnGen, an attention-guided training framework that embeds interpretability directly into the optimization process. AttnGen computes nucleotide-level importance scores using an attention mechanism and progressively suppresses low-contribution positions during training. This encourages the model to focus its predictions on a compact set of informative regions while reducing reliance on noisy sequence elements. We evaluate AttnGen on the standardized demo_human_or_worm benchmark, a binary classification task over 200-nucleotide sequences. With moderate masking, AttnGen achieves a validation accuracy of 96.73%, outperforming a conventional CNN baseline with 95.83% accuracy, while also exhibiting faster convergence and improved training stability. To assess whether the learned importance scores reflect functionally relevant signal, we conduct perturbation-based analysis by removing high-saliency nucleotides. This causes accuracy to drop from 96.9% to near chance level on a 3,000-sequence evaluation set, indicating that the model relies on a relatively small subset of informative positions. Our analysis shows that masking 10--20% of positions provides the most favorable trade-off between predictive performance and interpretability. These results suggest that attention-guided masking not only improves classification performance but also reshapes how models distribute importance across sequence positions. Although this study focuses on short genomic sequences, the proposed approach may extend to more complex interpretable sequence modeling settings.

Ali Karkehabadi

Deep Neural Networks are powerful tools to understand complex patterns and making decisions. However, their black-box nature impedes a complete understanding of their inner workings. While online saliency-guided training methods try to highlight the prominent features in the model's output to alleviate this problem, it is still ambiguous if the visually explainable features align with robustness of the model against adversarial examples. In this paper, we investigate the saliency trained model's vulnerability to adversarial examples methods. Models are trained using an online saliency-guided training method and evaluated against popular algorithms of adversarial examples. We quantify the robustness and conclude that despite the well-explained visualizations in the model's output, the salient models suffer from the lower performance against adversarial examples attacks.

Ali Karkehabadi, Houman Homayoun, Avesta Sasan

The Forward-Forward Learning (FFL) algorithm is a recently proposed solution for training neural networks without needing memory-intensive backpropagation. During training, labels accompany input data, classifying them as positive or negative inputs. Each layer learns its response to these inputs independently. In this study, we enhance the FFL with the following contributions: 1) We optimize label processing by segregating label and feature forwarding between layers, enhancing learning performance. 2) By revising label integration, we enhance the inference process, reduce computational complexity, and improve performance. 3) We introduce feedback loops akin to cortical loops in the brain, where information cycles through and returns to earlier neurons, enabling layers to combine complex features from previous layers with lower-level features, enhancing learning efficiency.

Mahla Ardebili Pour, Mohammad B. Ghiasi, Ali Karkehabadi

Floods are among the most prevalent and destructive natural disasters, often leading to severe social and economic impacts in urban areas due to the high concentration of assets and population density. In Iran, particularly in Tehran, recurring flood events underscore the urgent need for robust urban resilience strategies. This paper explores flood resilience models to identify the most effective approach for District 6 in Tehran. Through an extensive literature review, various resilience models were analyzed, with the Climate Disaster Resilience Index (CDRI) emerging as the most suitable model for this district due to its comprehensive resilience dimensions: Physical, Social, Economic, Organizational, and Natural Health resilience. Although the CDRI model provides a structured approach to resilience measurement, it remains a static model focused on spatial characteristics and lacks temporal adaptability. An extensive literature review enhances the CDRI model by integrating data from 2013 to 2022 in three-year intervals and applying machine learning techniques to predict resilience dimensions for 2025. This integration enables a dynamic resilience model that can accommodate temporal changes, providing a more adaptable and data driven foundation for urban flood resilience planning. By employing artificial intelligence to reflect evolving urban conditions, this model offers valuable insights for policymakers and urban planners to enhance flood resilience in Tehrans critical District 6.