Techniques for feedforward networks (FFNs) and convolutional networks (CNNs) are frequently reused across families, but the relationship between the underlying model classes is rarely made explicit. We introduce a unified node-level formalization with tensor-valued activations and show that generalized feedforward networks form a strict subset of generalized convolutional networks. Motivated by the mismatch in per-input parameterization between the two families, we propose model projection, a parameter-efficient transfer learning method for CNNs that freezes pretrained per-input-channel filters and learns a single scalar gate for each (output channel, input channel) contribution. Projection keeps all convolutional layers adaptable to downstream tasks while substantially reducing the number of trained parameters in convolutional layers. We prove that projected nodes take the generalized FFN form, enabling projected CNNs to inherit feedforward techniques that do not rely on homogeneous layer inputs. Experiments across multiple ImageNet-pretrained backbones and several downstream image classification datasets show that model projection is a strong transfer learning baseline under simple training recipes.

Transfer learning is often used in deep learning when data is limited, such as in medical imaging applications (Kim et al., 2022). Foundation models, that is large, publicly available, pre-trained models, are often fine-tuned for such tasks where little data is available (Wang et al., 2023; Zhang and Metaxas, 2024; Khan et al., 2025). Beyond freezing part of a model to reduce overfitting, various techniques can increase training data such as data augmentation, and self supervised learning. These methods can reduce overfitting (Chollet, 2021; Wang et al., 2023; Ewen and Khan, 2021), but still struggle when there is little data available (Wang et al., 2023). We propose a new approach, Structured Output Regularization (SOR), a simple framework that adapts and prunes pretrained networks using very little labeled data. Instead of unfreezing internal weights, SOR keeps internal structures frozen, e.g., convolutional filters or higher-level blocks, and regularizes their outputs. Specifically, we freeze internal structure weights, we add new weights between each frozen structure, penalized via lasso penalty to encourage sparsity, and train the network. Structures whose new weights are driven to zero can be removed, yielding a smaller, task-tailored model without training the full parameter set. To regularize the final layer structures, SOR applies group lasso.

Attempt to use convolutional neural network to achieve kinematic analysis of plane bar structure. Through 3dsMax animation software and OpenCV module, self-build image dataset of geometrically stable system and geometrically unstable system. we construct and train convolutional neural network model based on the TensorFlow and Keras deep learning platform framework. The model achieves 100% accuracy on the training set, validation set, and test set. The accuracy on the additional test set is 93.7%, indicating that convolutional neural network can learn and master the relevant knowledge of kinematic analysis of structural mechanics. In the future, the generalization ability of the model can be improved through the diversity of dataset, which has the potential to surpass human experts for complex structures. Convolutional neural network has certain practical value in the field of kinematic analysis of structural mechanics. Using visualization technology, we reveal how convolutional neural network learns and recognizes structural features. Using pre-trained VGG16 model for feature extraction and fine-tuning, we found that the generalization ability is inferior to the self-built model.

Serial femtosecond crystallography at X-ray free electron laser facilities opens a new era for the determination of crystal structure. However, the data processing of those experiments is facing unprecedented challenge, because the total number of diffraction patterns needed to determinate a high-resolution structure is huge. Machine learning methods are very likely to play important roles in dealing with such a large volume of data. Convolutional neural networks have made a great success in the field of pattern classification, however, training of the networks need very large datasets with labels. Th is heavy dependence on labeled datasets will seriously restrict the application of networks, because it is very costly to annotate a large number of diffraction patterns. In this article we present our job on the classification of diffraction pattern by weakly supervised algorithms, with the aim of reducing as much as possible the size of the labeled dataset required for training. Our result shows that weakly supervised methods can significantly reduce the need for the number of labeled patterns while achieving comparable accuracy to fully supervised methods.

Most recent developments in AI, including computer vision, natural language processing, predictive analytics, autonomous systems, and a wide range of applications, are driven by machine learning. We need data so these algorithms can learn from them so can generalize well. Most of the transitional algorithms need label data, to work. When it comes to deep learning, the amount the data requires is humongous, particularly in deep learning neural networks, compared to traditional machine learning algorithms, to build a model that achieves the appropriate levels of accuracy. Therefore, it should go without saying that for the resulting machine-learning models to be accurate, the machine-learning data must be clean, accurate, full, and well-labeled.

Creating intelligent systems capable of recognizing emotions is a difficult task, especially when looking at emotions in animals. This paper describes the process of designing a "proof of concept" system to recognize emotions in horses. This system is formed by two elements, a detector and a model. The detector is a fast region-based convolutional neural network that detects horses in an image. The model is a convolutional neural network that predicts the emotions of those horses. These two elements were trained with multiple images of horses until they achieved high accuracy in their tasks. In total, 400 images of horses were collected and labeled to train both the detector and the model while 40 were used to test the system. Once the two components were validated, they were combined into a testable system that would detect equine emotions based on established behavioral ethograms indicating emotional affect through head, neck, ear, muzzle and eye position. The system showed an accuracy of 80% on the validation set and 65% on the test set, demonstrating that it is possible to predict emotions in animals using autonomous intelligent systems. Such a system has multiple applications including further studies in the growing field of animal emotions as well as in the veterinary field to determine the physical welfare of horses or other livestock.

Image and video-capturing technologies have permeated our every-day life. Such technologies can continuously monitor individuals' expressions in real-life settings, affording us new insights into their emotional states and transitions, thus paving the way to novel well-being and healthcare applications. Yet, due to the strong privacy concerns, the use of such technologies is met with strong skepticism, since current face-based emotion recognition systems relying on deep learning techniques tend to preserve substantial information related to the identity of the user, apart from the emotion-specific information. This paper proposes an adversarial learning framework which relies on a convolutional neural network (CNN) architecture trained through an iterative procedure for minimizing identity-specific information and maximizing emotion-dependent information. The proposed approach is evaluated through emotion classification and face identification metrics, and is compared against two CNNs, one trained solely for emotion recognition and the other trained solely for face identification. Experiments are performed using the Yale Face Dataset and Japanese Female Facial Expression Database. Results indicate that the proposed approach can learn a convolutional transformation for preserving emotion recognition accuracy and degrading face identity recognition, providing a foundation toward privacy-aware emotion recognition technologies.

This last point can be addressed by using the tf.data.Dataset.apply()

Read it now to have an idea why we do what we do here. See you just happen to be in a region where you do not have access to any Google's websites, which kindly sucks, so I summarized it here for you. You should use Dataset API to create input pipelines for TensorFlow models.

Read it now to have an idea why we do what we do here. See you just happen to be in a region where you do not have access to any Google's websites, which kindly sucks, so I summarized it here for you. You should use Dataset API to create input pipelines for TensorFlow models.