Explaining neural scaling laws.

The population loss of trained deep neural networks often follows precise power-law scaling relations with either the size of the training dataset or the number of parameters in the network. We propose a theory that explains the origins of and connects these scaling laws. We identify variance-limited and resolution-limited scaling behavior for both dataset and model size, for a total of four scaling regimes. The variance-limited scaling follows simply from the existence of a well-behaved infinite data or infinite width limit, while the resolution-limited regime can be explained by positing that models are effectively resolving a smooth data manifold. In the large width limit, this can be equivalently obtained from the spectrum of certain kernels, and we present evidence that large width and large dataset resolution-limited scaling exponents are related by a duality. We exhibit all four scaling regimes in the controlled setting of large random feature and pretrained models and test the predictions empirically on a range of standard architectures and datasets. We also observe several empirical relationships between datasets and scaling exponents under modifications of task and architecture aspect ratio. Our work provides a taxonomy for classifying different scaling regimes, underscores that there can be different mechanisms driving improvements in loss, and lends insight into the microscopic origin and relationships between scaling exponents.

Representations and generalization in artificial and brain neural networks.

Humans and animals excel at generalizing from limited data, a capability yet to be fully replicated in artificial intelligence. This perspective investigates generalization in biological and artificial deep neural networks (DNNs), in both in-distribution and out-of-distribution contexts. We introduce two hypotheses: First, the geometric properties of the neural manifolds associated with discrete cognitive entities, such as objects, words, and concepts, are powerful order parameters. They link the neural substrate to the generalization capabilities and provide a unified methodology bridging gaps between neuroscience, machine learning, and cognitive science. We overview recent progress in studying the geometry of neural manifolds, particularly in visual object recognition, and discuss theories connecting manifold dimension and radius to generalization capacity. Second, we suggest that the theory of learning in wide DNNs, especially in the thermodynamic limit, provides mechanistic insights into the learning processes generating desired neural representational geometries and generalization. This includes the role of weight norm regularization, network architecture, and hyper-parameters. We will explore recent advances in this theory and ongoing challenges. We also discuss the dynamics of learning and its relevance to the issue of representational drift in the brain.

Likelihood-based feature representation learning combined with neighborhood information for predicting circRNA-miRNA associations.

Connections between circular RNAs (circRNAs) and microRNAs (miRNAs) assume a pivotal position in the onset, evolution, diagnosis and treatment of diseases and tumors. Selecting the most potential circRNA-related miRNAs and taking advantage of them as the biological markers or drug targets could be conducive to dealing with complex human diseases through preventive strategies, diagnostic procedures and therapeutic approaches. Compared to traditional biological experiments, leveraging computational models to integrate diverse biological data in order to infer potential associations proves to be a more efficient and cost-effective approach. This paper developed a model of Convolutional Autoencoder for CircRNA-MiRNA Associations (CA-CMA) prediction. Initially, this model merged the natural language characteristics of the circRNA and miRNA sequence with the features of circRNA-miRNA interactions. Subsequently, it utilized all circRNA-miRNA pairs to construct a molecular association network, which was then fine-tuned by labeled samples to optimize the network parameters. Finally, the prediction outcome is obtained by utilizing the deep neural networks classifier. This model innovatively combines the likelihood objective that preserves the neighborhood through optimization, to learn the continuous feature representation of words and preserve the spatial information of two-dimensional signals. During the process of 5-fold cross-validation, CA-CMA exhibited exceptional performance compared to numerous prior computational approaches, as evidenced by its mean area under the receiver operating characteristic curve of 0.9138 and a minimal SD of 0.0024. Furthermore, recent literature has confirmed the accuracy of 25 out of the top 30 circRNA-miRNA pairs identified with the highest CA-CMA scores during case studies. The results of these experiments highlight the robustness and versatility of our model.

Behavioral signatures of face perception emerge in deep neural networks optimized for face recognition.

Human face recognition is highly accurate and exhibits a number of distinctive and well-documented behavioral "signatures" such as the use of a characteristic representational space, the disproportionate performance cost when stimuli are presented upside down, and the drop in accuracy for faces from races the participant is less familiar with. These and other phenomena have long been taken as evidence that face recognition is "special". But why does human face perception exhibit these properties in the first place? Here, we use deep convolutional neural networks (CNNs) to test the hypothesis that all of these signatures of human face perception result from optimization for the task of face recognition. Indeed, as predicted by this hypothesis, these phenomena are all found in CNNs trained on face recognition, but not in CNNs trained on object recognition, even when additionally trained to detect faces while matching the amount of face experience. To test whether these signatures are in principle specific to faces, we optimized a CNN on car discrimination and tested it on upright and inverted car images. As we found for face perception, the car-trained network showed a drop in performance for inverted vs. upright cars. Similarly, CNNs trained on inverted faces produced an inverted face inversion effect. These findings show that the behavioral signatures of human face perception reflect and are well explained as the result of optimization for the task of face recognition, and that the nature of the computations underlying this task may not be so special after all.

Sensing prior constraints in deep neural networks for solving exploration geophysical problems.

One of the key objectives in geophysics is to characterize the subsurface through the process of analyzing and interpreting geophysical field data that are typically acquired at the surface. Data-driven deep learning methods have enormous potential for accelerating and simplifying the process but also face many challenges, including poor generalizability, weak interpretability, and physical inconsistency. We present three strategies for imposing domain knowledge constraints on deep neural networks (DNNs) to help address these challenges. The first strategy is to integrate constraints into data by generating synthetic training datasets through geological and geophysical forward modeling and properly encoding prior knowledge as part of the input fed into the DNNs. The second strategy is to design nontrainable custom layers of physical operators and preconditioners in the DNN architecture to modify or shape feature maps calculated within the network to make them consistent with the prior knowledge. The final strategy is to implement prior geological information and geophysical laws as regularization terms in loss functions for training the DNNs. We discuss the implementation of these strategies in detail and demonstrate their effectiveness by applying them to geophysical data processing, imaging, interpretation, and subsurface model building.

DP2LM: leveraging deep learning approach for estimation and hypothesis testing on mediation effects with high-dimensional mediators and complex confounders.

Traditional linear mediation analysis has inherent limitations when it comes to handling high-dimensional mediators. Particularly, accurately estimating and rigorously inferring mediation effects is challenging, primarily due to the intertwined nature of the mediator selection issue. Despite recent developments, the existing methods are inadequate for addressing the complex relationships introduced by confounders. To tackle these challenges, we propose a novel approach called DP2LM (Deep neural network-based Penalized Partially Linear Mediation). This approach incorporates deep neural network techniques to account for nonlinear effects in confounders and utilizes the penalized partially linear model to accommodate high dimensionality. Unlike most existing works that concentrate on mediator selection, our method prioritizes estimation and inference on mediation effects. Specifically, we develop test procedures for testing the direct and indirect mediation effects. Theoretical analysis shows that the tests maintain the Type-I error rate. In simulation studies, DP2LM demonstrates its superior performance as a modeling tool for complex data, outperforming existing approaches in a wide range of settings and providing reliable estimation and inference in scenarios involving a considerable number of mediators. Further, we apply DP2LM to investigate the mediation effect of DNA methylation on cortisol stress reactivity in individuals who experienced childhood trauma, uncovering new insights through a comprehensive analysis.

Deep autoregressive generative models capture the intrinsics embedded in T-cell receptor repertoires.

T-cell receptors (TCRs) play an essential role in the adaptive immune system. Probabilistic models for TCR repertoires can help decipher the underlying complex sequence patterns and provide novel insights into understanding the adaptive immune system. In this work, we develop TCRpeg, a deep autoregressive generative model to unravel the sequence patterns of TCR repertoires. TCRpeg largely outperforms state-of-the-art methods in estimating the probability distribution of a TCR repertoire, boosting the average accuracy from 0.672 to 0.906 measured by the Pearson correlation coefficient. Furthermore, with promising performance in probability inference, TCRpeg improves on a range of TCR-related tasks: profiling TCR repertoire probabilistically, classifying antigen-specific TCRs, validating previously discovered TCR motifs, generating novel TCRs and augmenting TCR data. Our results and analysis highlight the flexibility and capacity of TCRpeg to extract TCR sequence information, providing a novel approach for deciphering complex immunogenomic repertoires.

Quantifying the narrative flow of imagined versus autobiographical stories.

Lifelong experiences and learned knowledge lead to shared expectations about how common situations tend to unfold. Such knowledge of narrative event flow enables people to weave together a story. However, comparable computational tools to evaluate the flow of events in narratives are limited. We quantify the differences between autobiographical and imagined stories by introducing sequentiality, a measure of narrative flow of events, drawing probabilistic inferences from a cutting-edge large language model (GPT-3). Sequentiality captures the flow of a narrative by comparing the probability of a sentence with and without its preceding story context. We applied our measure to study thousands of diary-like stories, collected from crowdworkers, about either a recent remembered experience or an imagined story on the same topic. The results show that imagined stories have higher sequentiality than autobiographical stories and that the sequentiality of autobiographical stories increases when the memories are retold several months later. In pursuit of deeper understandings of how sequentiality measures the flow of narratives, we explore proportions of major and minor events in story sentences, as annotated by crowdworkers. We find that lower sequentiality is associated with higher proportions of major events. The methods and results highlight opportunities to use cutting-edge computational analyses, such as sequentiality, on large corpora of matched imagined and autobiographical stories to investigate the influences of memory and reasoning on language generation processes.

Face dissimilarity judgments are predicted by representational distance in morphable and image-computable models.

Human vision is attuned to the subtle differences between individual faces. Yet we lack a quantitative way of predicting how similar two face images look and whether they appear to show the same person. Principal component-based three-dimensional (3D) morphable models are widely used to generate stimuli in face perception research. These models capture the distribution of real human faces in terms of dimensions of physical shape and texture. How well does a "face space" based on these dimensions capture the similarity relationships humans perceive among faces? To answer this, we designed a behavioral task to collect dissimilarity and same/different identity judgments for 232 pairs of realistic faces. Stimuli sampled geometric relationships in a face space derived from principal components of 3D shape and texture (Basel face model [BFM]). We then compared a wide range of models in their ability to predict the data, including the BFM from which faces were generated, an active appearance model derived from face photographs, and image-computable models of visual perception. Euclidean distance in the BFM explained both dissimilarity and identity judgments surprisingly well. In a comparison against 16 diverse models, BFM distance was competitive with representational distances in state-of-the-art deep neural networks (DNNs), including novel DNNs trained on BFM synthetic identities or BFM latents. Models capturing the distribution of face shape and texture across individuals are not only useful tools for stimulus generation. They also capture important information about how faces are perceived, suggesting that human face representations are tuned to the statistical distribution of faces.

Texture-like representation of objects in human visual cortex.

The human visual ability to recognize objects and scenes is widely thought to rely on representations in category-selective regions of the visual cortex. These representations could support object vision by specifically representing objects, or, more simply, by representing complex visual features regardless of the particular spatial arrangement needed to constitute real-world objects, that is, by representing visual textures. To discriminate between these hypotheses, we leveraged an image synthesis approach that, unlike previous methods, provides independent control over the complexity and spatial arrangement of visual features. We found that human observers could easily detect a natural object among synthetic images with similar complex features that were spatially scrambled. However, observer models built from BOLD responses from category-selective regions, as well as a model of macaque inferotemporal cortex and Imagenet-trained deep convolutional neural networks, were all unable to identify the real object. This inability was not due to a lack of signal to noise, as all observer models could predict human performance in image categorization tasks. How then might these texture-like representations in category-selective regions support object perception? An image-specific readout from category-selective cortex yielded a representation that was more selective for natural feature arrangement, showing that the information necessary for natural object discrimination is available. Thus, our results suggest that the role of the human category-selective visual cortex is not to explicitly encode objects but rather to provide a basis set of texture-like features that can be infinitely reconfigured to flexibly learn and identify new object categories.

Deep Neural Networks and Visuo-Semantic Models Explain Complementary Components of Human Ventral-Stream Representational Dynamics.

Deep neural networks (DNNs) are promising models of the cortical computations supporting human object recognition. However, despite their ability to explain a significant portion of variance in neural data, the agreement between models and brain representational dynamics is far from perfect. We address this issue by asking which representational features are currently unaccounted for in neural time series data, estimated for multiple areas of the ventral stream via source-reconstructed magnetoencephalography data acquired in human participants (nine females, six males) during object viewing. We focus on the ability of visuo-semantic models, consisting of human-generated labels of object features and categories, to explain variance beyond the explanatory power of DNNs alone. We report a gradual reversal in the relative importance of DNN versus visuo-semantic features as ventral-stream object representations unfold over space and time. Although lower-level visual areas are better explained by DNN features starting early in time (at 66 ms after stimulus onset), higher-level cortical dynamics are best accounted for by visuo-semantic features starting later in time (at 146 ms after stimulus onset). Among the visuo-semantic features, object parts and basic categories drive the advantage over DNNs. These results show that a significant component of the variance unexplained by DNNs in higher-level cortical dynamics is structured and can be explained by readily nameable aspects of the objects. We conclude that current DNNs fail to fully capture dynamic representations in higher-level human visual cortex and suggest a path toward more accurate models of ventral-stream computations.SIGNIFICANCE STATEMENT When we view objects such as faces and cars in our visual environment, their neural representations dynamically unfold over time at a millisecond scale. These dynamics reflect the cortical computations that support fast and robust object recognition. DNNs have emerged as a promising framework for modeling these computations but cannot yet fully account for the neural dynamics. Using magnetoencephalography data acquired in human observers during object viewing, we show that readily nameable aspects of objects, such as 'eye', 'wheel', and 'face', can account for variance in the neural dynamics over and above DNNs. These findings suggest that DNNs and humans may in part rely on different object features for visual recognition and provide guidelines for model improvement.

Intracranial Electroencephalography and Deep Neural Networks Reveal Shared Substrates for Representations of Face Identity and Expressions.

According to a classical view of face perception (Bruce and Young, 1986; Haxby et al., 2000), face identity and facial expression recognition are performed by separate neural substrates (ventral and lateral temporal face-selective regions, respectively). However, recent studies challenge this view, showing that expression valence can also be decoded from ventral regions (Skerry and Saxe, 2014; Li et al., 2019), and identity from lateral regions (Anzellotti and Caramazza, 2017). These findings could be reconciled with the classical view if regions specialized for one task (either identity or expression) contain a small amount of information for the other task (that enables above-chance decoding). In this case, we would expect representations in lateral regions to be more similar to representations in deep convolutional neural networks (DCNNs) trained to recognize facial expression than to representations in DCNNs trained to recognize face identity (the converse should hold for ventral regions). We tested this hypothesis by analyzing neural responses to faces varying in identity and expression. Representational dissimilarity matrices (RDMs) computed from human intracranial recordings (n = 11 adults; 7 females) were compared with RDMs from DCNNs trained to label either identity or expression. We found that RDMs from DCNNs trained to recognize identity correlated with intracranial recordings more strongly in all regions tested-even in regions classically hypothesized to be specialized for expression. These results deviate from the classical view, suggesting that face-selective ventral and lateral regions contribute to the representation of both identity and expression.SIGNIFICANCE STATEMENT Previous work proposed that separate brain regions are specialized for the recognition of face identity and facial expression. However, identity and expression recognition mechanisms might share common brain regions instead. We tested these alternatives using deep neural networks and intracranial recordings from face-selective brain regions. Deep neural networks trained to recognize identity and networks trained to recognize expression learned representations that correlate with neural recordings. Identity-trained representations correlated with intracranial recordings more strongly in all regions tested, including regions hypothesized to be expression specialized in the classical hypothesis. These findings support the view that identity and expression recognition rely on common brain regions. This discovery may require reevaluation of the roles that the ventral and lateral neural pathways play in processing socially relevant stimuli.

Category Trumps Shape as an Organizational Principle of Object Space in the Human Occipitotemporal Cortex.

The organizational principles of the object space represented in the human ventral visual cortex are debated. Here we contrast two prominent proposals that, in addition to an organization in terms of animacy, propose either a representation related to aspect ratio (stubby-spiky) or to the distinction between faces and bodies. We designed a critical test that dissociates the latter two categories from aspect ratio and investigated responses from human fMRI (of either sex) and deep neural networks (BigBiGAN). Representational similarity and decoding analyses showed that the object space in the occipitotemporal cortex and BigBiGAN was partially explained by animacy but not by aspect ratio. Data-driven approaches showed clusters for face and body stimuli and animate-inanimate separation in the representational space of occipitotemporal cortex and BigBiGAN, but no arrangement related to aspect ratio. In sum, the findings go in favor of a model in terms of an animacy representation combined with strong selectivity for faces and bodies.SIGNIFICANCE STATEMENT We contrasted animacy, aspect ratio, and face-body as principal dimensions characterizing object space in the occipitotemporal cortex. This is difficult to test, as typically faces and bodies differ in aspect ratio (faces are mostly stubby and bodies are mostly spiky). To dissociate the face-body distinction from the difference in aspect ratio, we created a new stimulus set in which faces and bodies have a similar and very wide distribution of values along the shape dimension of the aspect ratio. Brain imaging (fMRI) with this new stimulus set showed that, in addition to animacy, the object space is mainly organized by the face-body distinction and selectivity for aspect ratio is minor (despite its wide distribution).

Prediction of anticancer drug sensitivity using an interpretable model guided by deep learning.

BACKGROUND: The prediction of drug sensitivity plays a crucial role in improving the therapeutic effect of drugs. However, testing the effectiveness of drugs is challenging due to the complex mechanism of drug reactions and the lack of interpretability in most machine learning and deep learning methods. Therefore, it is imperative to establish an interpretable model that receives various cell line and drug feature data to learn drug response mechanisms and achieve stable predictions between available datasets. RESULTS: This study proposes a new and interpretable deep learning model, DrugGene, which integrates gene expression, gene mutation, gene copy number variation of cancer cells, and chemical characteristics of anticancer drugs to predict their sensitivity. This model comprises two different branches of neural networks, where the first involves a hierarchical structure of biological subsystems that uses the biological processes of human cells to form a visual neural network (VNN) and an interpretable deep neural network for human cancer cells. DrugGene receives genotype input from the cell line and detects changes in the subsystem states. We also employ a traditional artificial neural network (ANN) to capture the chemical structural features of drugs. DrugGene generates final drug response predictions by combining VNN and ANN and integrating their outputs into a fully connected layer. The experimental results using drug sensitivity data extracted from the Cancer Drug Sensitivity Genome Database and the Cancer Treatment Response Portal v2 reveal that the proposed model is better than existing prediction methods. Therefore, our model achieves higher accuracy, learns the reaction mechanisms between anticancer drugs and cell lines from various features, and interprets the model's predicted results. CONCLUSIONS: Our method utilizes biological pathways to construct neural networks, which can use genotypes to monitor changes in the state of network subsystems, thereby interpreting the prediction results in the model and achieving satisfactory prediction accuracy. This will help explore new directions in cancer treatment. More available code resources can be downloaded for free from GitHub ( https://github.com/pangweixiong/DrugGene ).

Learning self-supervised molecular representations for drug-drug interaction prediction.

Drug-drug interactions (DDI) are a critical concern in healthcare due to their potential to cause adverse effects and compromise patient safety. Supervised machine learning models for DDI prediction need to be optimized to learn abstract, transferable features, and generalize to larger chemical spaces, primarily due to the scarcity of high-quality labeled DDI data. Inspired by recent advances in computer vision, we present SMR-DDI, a self-supervised framework that leverages contrastive learning to embed drugs into a scaffold-based feature space. Molecular scaffolds represent the core structural motifs that drive pharmacological activities, making them valuable for learning informative representations. Specifically, we pre-trained SMR-DDI on a large-scale unlabeled molecular dataset. We generated augmented views for each molecule via SMILES enumeration and optimized the embedding process through contrastive loss minimization between views. This enables the model to capture relevant and robust molecular features while reducing noise. We then transfer the learned representations for the downstream prediction of DDI. Experiments show that the new feature space has comparable expressivity to state-of-the-art molecular representations and achieved competitive DDI prediction results while training on less data. Additional investigations also revealed that pre-training on more extensive and diverse unlabeled molecular datasets improved the model's capability to embed molecules more effectively. Our results highlight contrastive learning as a promising approach for DDI prediction that can identify potentially hazardous drug combinations using only structural information.

Deep neural network models for cell type prediction based on single-cell Hi-C data.

BACKGROUND: Cell type prediction is crucial to cell type identification of genomics, cancer diagnosis and drug development, and it can solve the time-consuming and difficult problem of cell classification in biological experiments. Therefore, a computational method is urgently needed to classify and predict cell types using single-cell Hi-C data. In previous studies, there is a lack of convenient and accurate method to predict cell types based on single-cell Hi-C data. Deep neural networks can form complex representations of single-cell Hi-C data and make it possible to handle the multidimensional and sparse biological datasets. RESULTS: We compare the performance of SCANN with existing methods and analyze the model by using five different evaluation metrics. When using only ML1 and ML3 datasets, the ARI and NMI values of SCANN increase by 14% and 11% over those of scHiCluster respectively. However, when using all six libraries of data, the ARI and NMI values of SCANN increase by 63% and 88% over those of scHiCluster respectively. These findings show that SCANN is highly accurate in predicting the type of independent cell samples using single-cell Hi-C data. CONCLUSIONS: SCANN enhances the training speed and requires fewer resources for predicting cell types. In addition, when the number of cells in different cell types was extremely unbalanced, SCANN has higher stability and flexibility in solving cell classification and cell type prediction using the single-cell Hi-C data. This predication method can assist biologists to study the differences in the chromosome structure of cells between different cell types.

Discovering individual fingerprints in resting-state functional connectivity using deep neural networks.

Non-negligible idiosyncrasy due to interindividual differences is an ongoing issue in resting-state functional MRI (rfMRI) analysis. We show that a deep neural network (DNN) can be employed for individual identification by learning important features from the time-varying functional connectivity (FC) of rfMRI in the Human Connectome Project. We employed the trained DNN to identify individuals from an independent dataset acquired at our institution. The results revealed that the DNN could successfully identify 300 individuals with an error rate of 2.9% using 15 s time-window and 870 individuals with an error rate of 6.7%. A trained DNN with nonlinear hidden layers led to the proposal of the "fingerprint of FC" (fpFC) as representative edges of individual FC. The fpFCs for individuals exhibited commonly important and individual-specific edges across time-window lengths (from 5 min to 15 s). Furthermore, the utility of our model for another group of subjects was validated, supporting the feasibility of our technique in the context of transfer learning. In conclusion, our study offers an insight into the discovery of the intrinsic mode of the human brain using whole-brain resting-state FC and DNNs.

Multimodal deep learning for biomedical data fusion: a review.

Biomedical data are becoming increasingly multimodal and thereby capture the underlying complex relationships among biological processes. Deep learning (DL)-based data fusion strategies are a popular approach for modeling these nonlinear relationships. Therefore, we review the current state-of-the-art of such methods and propose a detailed taxonomy that facilitates more informed choices of fusion strategies for biomedical applications, as well as research on novel methods. By doing so, we find that deep fusion strategies often outperform unimodal and shallow approaches. Additionally, the proposed subcategories of fusion strategies show different advantages and drawbacks. The review of current methods has shown that, especially for intermediate fusion strategies, joint representation learning is the preferred approach as it effectively models the complex interactions of different levels of biological organization. Finally, we note that gradual fusion, based on prior biological knowledge or on search strategies, is a promising future research path. Similarly, utilizing transfer learning might overcome sample size limitations of multimodal data sets. As these data sets become increasingly available, multimodal DL approaches present the opportunity to train holistic models that can learn the complex regulatory dynamics behind health and disease.

Clair3-trio: high-performance Nanopore long-read variant calling in family trios with trio-to-trio deep neural networks.

Accurate identification of genetic variants from family child-mother-father trio sequencing data is important in genomics. However, state-of-the-art approaches treat variant calling from trios as three independent tasks, which limits their calling accuracy for Nanopore long-read sequencing data. For better trio variant calling, we introduce Clair3-Trio, the first variant caller tailored for family trio data from Nanopore long-reads. Clair3-Trio employs a Trio-to-Trio deep neural network model, which allows it to input the trio sequencing information and output all of the trio's predicted variants within a single model to improve variant calling. We also present MCVLoss, a novel loss function tailor-made for variant calling in trios, leveraging the explicit encoding of the Mendelian inheritance. Clair3-Trio showed comprehensive improvement in experiments. It predicted far fewer Mendelian inheritance violation variations than current state-of-the-art methods. We also demonstrated that our Trio-to-Trio model is more accurate than competing architectures. Clair3-Trio is accessible as a free, open-source project at https://github.com/HKU-BAL/Clair3-Trio.

sAMP-PFPDeep: Improving accuracy of short antimicrobial peptides prediction using three different sequence encodings and deep neural networks.

Short antimicrobial peptides (sAMPs) belong to a significant repertoire of antimicrobial agents and are known to possess enhanced antimicrobial activity, higher stability and less toxicity to human cells, as well as less complex than other large biological drugs. As these molecules are significantly important, herein, a prediction method for sAMPs (with a sequence length ≤ 30 residues) is proposed for accurate and efficient prediction of sAMPs instead of laborious and costly experimental approaches. Benchmark dataset was collected from a recently reported study and sequences were converted into three channel images comprising information related to the position, frequency and sum of 12 physiochemical features as the first, second and third channels, respectively. Two image-based deep neural networks (DNNs), i.e. RESNET-50 and VGG-16 were trained and evaluated using various metrics while a comparative analysis with previous techniques was also performed. Validation of sAMP-PFPDeep was also performed by using molecular docking based analysis. The results showed that VGG-16 provided more accurate results, i.e. 98.30% training accuracy and 87.37% testing accuracy for predicting sAMPs as compared to those of RESNET-50 having 96.14% training accuracy and 83.87% testing accuracy. However, the comparative analysis revealed that both these models outperformed previously reported state-of-the-art methods. Based on the results, it is concluded that sAMP-PFPDeep can help identify antimicrobial peptides with promising accuracy and efficiency. It can help biologists and scientists to identify antimicrobial peptides, by further aiding the computer-aided drug design and discovery, as well as virtual screening protocols against various pathologies. sAMP-PFPDeep is available at (https://github.com/WaqarHusain/sAMP-PFPDeep).