This article discusses MusGConv, a perception-inspired graph convolution block for symbolic musical applications 10 min read

In the field of Music Information Research (MIR), the challenge of understanding and processing musical scores has continuously been introduced to new methods and approaches. Most recently many graph-based techniques have been proposed as a way to target music understanding tasks such as voice separation, cadence detection, composer classification, and Roman numeral analysis.

This blog post covers one of my recent papers in which I introduced a new graph convolutional block, called MusGConv, designed specifically for processing music score data. MusGConv takes advantage of music perceptual principles to improve the efficiency and the performance of graph convolution in Graph Neural Networks applied to music understanding tasks.

Traditional approaches in MIR often rely on audio or symbolic representations of music. While audio captures the intensity of sound waves over time, symbolic representations like MIDI files or musical scores encode discrete musical events. Symbolic representations are particularly valuable as they provide higher-level information essential for tasks such as music analysis and generation.

However, existing techniques based on symbolic music representations often borrow from computer vision (CV) or natural language processing (NLP) methodologies. For instance, representing music as a pianoroll in a matrix format and treating it similarly to an image, or, representing music as a series of tokens and treating it with sequential models or transformers. These approaches, though effective, could fall short in fully capturing the complex, multi-dimensional nature of music, which includes hierarchical note relation and intricate pitch-temporal relationships. Some recent approaches have been proposed to model the musical score as a graph and apply Graph Neural Networks to solve various tasks.

The fundamental idea of GNN-based approaches to musical scores is to model a musical score as a graph where notes are the vertices and edges are built from the temporal relations between the notes. To create a graph from a musical score we can consider four types of edges (see Figure below for a visualization of the graph on the score):

A GNN can treat the graph created from the notes and these four types of relations.

MusGConv is designed to leverage music score graphs and enhance them by incorporating principles of music perception into the graph convolution process. It focuses on two fundamental dimensions of music: pitch and rhythm, considering both their relative and absolute representations.

Absolute representations refer to features that can be attributed to each note individually such as the notes pitch or spelling, its duration or any other feature. On the other hand, relative features are computed between pairs of notes, such as the music interval between two notes, their onset difference, i.e. the time on which they occur, etc.

The importance and coexistence of the relative and absolute representations can be understood from a transpositional perspective in music. Imagine the same music content transposed. Then, the intervalic relations between notes stay the same but the pitch of each note is altered.

To fully understand the inner workings of the MusGConv convolution block it is important to first explain the principles of Message Passing.

In the context of GNNs, message passing is a process where vertices within a graph exchange information with their neighbors to update their own representations. This exchange allows each node to gather contextual information from the graph, which is then used to for predictive tasks.

The message passing process is defined by the following steps:

MusGConv alters the standard message passing process mainly by incorporating both absolute features as node features and relative musical features as edge features. This design is tailored to fit the nature of musical data.

The MusGConv convolution is defined by the following steps:

By designing the message passing mechanism in this way, MusGConv attempts to preserve the relative perceptual properties of music (such as intervals and rhythms), leading to more meaningful representations of musical data.

Should edge features are absent or deliberately not provided then MusGConv computes the edge features between two nodes as the absolute difference between their node features. The version of MusGConv with the edges features is named MusGConv(+EF) in the experiments.

To demonstrate the potential of MusGConv I discuss below the tasks and the experiments conducted in the paper. All models independent of the task are designed with the pipeline shown in the figure below. When MusGConv is employed the GNN blocks are replaced by MusGConv blocks.

I decided to apply MusGConv to four tasks: voice separation, composer classification, Roman numeral analysis, and cadence detection. Each one of these tasks presents a different taxonomy from a graph learning perspective. Voice separation is a link prediction task, composer classification is a global classification task, cadence detection is a node classification task, and Roman numeral analysis can be viewed as a subgraph classification task. Therefore we are exploring the suitability of MusGConv not only from a musical analysis perspective but through out the spectrum of graph deep learning task taxonomy.

Read the original:

Perception-Inspired Graph Convolution for Music Understanding Tasks | by Emmanouil Karystinaios | Jul, 2024 - Towards Data Science

When fictions most famous detective, Sherlock Holmes, needed to solve a crime, he turned to his sharp observational skills and deep understanding of human nature. He used this combination more than once when facing off against his arch-nemesis, Dr James Moriarty, a villain adept at exploiting human weaknesses for his gain.

This classic battle mirrors todays ongoing fight against cybercrime. Like Moriarty, cybercriminals use cunning strategies to exploit their victims psychological vulnerabilities. They send deceptive emails or messages that appear to be from trusted sources such as banks, employers, or friends. These messages often contain urgent requests or alarming information to provoke an immediate response.

For example, a phishing email might claim there has been suspicious activity on a victims bank account and prompt them to click on a link to verify their account details. Once the victim clicks the link and enters their information, the attackers capture their credentials for malicious use. Or individuals are manipulated into divulging confidential information to compromise their own or a companys security.

Holmes had to outsmart Moriarty by understanding and anticipating his moves. Modern cybersecurity teams and users must stay vigilant and proactive to outmanoeuvre cybercriminals who continuously refine their deceptive tactics.

Read more: Deepfakes in South Africa: protecting your image online is the key to fighting them

What if those trying to prevent cybercrime could harness Holmess skills? Could those skills complement existing, more data-driven ways of identifying potential threats? I am a professor of information systems whose research focuses on, among other things, integrating data science and behavioural science through a sociotechnical lens to investigate the deceptive tactics used by cybercriminals.

Recently, I worked with Shiven Naidoo, a Masters student in data science, to understand how behavioural science and data science could join forces to combat cybercrime.

Our study found that, just as Holmess analytical genius and his sidekick Dr John Watsons practical approach were complementary, behavioural scientists and data scientists can collaborate to make cybercrime detection and prevention models more effective.

Data science uses scientific methods, processes, algorithms and systems to extract knowledge and insights from structured and unstructured data.

When its powerful algorithms are applied to complex and large datasets they can identify patterns that indicate potential cyber threats. Predictive analysis helps cybersecurity teams anticipate and prevent large-scale attacks. This is done through, for instance, detecting anomalies in sentence structure to spot scams.

However, relying solely on data science often overlooks the human factors that drive cybercriminal behaviour.

The behavioural sciences study human behaviour. They consider the principles that influence decision-making and compliance. We drew extensively from US psychologist Robert Cialdinis social influence model in our study.

This model has been applied in cybersecurity studies to explain how cybercriminals exploit psychological tendencies.

Read more: Five things South Africa must do to combat cybercrime

For example, cybercriminals exploit humans tendency to be obedient to authority by impersonating trusted figures to spread disinformation. They also exploit urgency and scarcity principles to prompt hasty actions. Social proof the tendency to follow the actions of those similar to us is another tool, used to manipulate users into complying with fraudulent requests. For instance, cybercriminals might create fake reviews or testimonials, prompting users to fall for a scam.

We adapted the social influence model to detect cybercriminal tactics in scam datasets by combining behavioural and data science. Scam datasets consist of unstructured data, which includes complex text data such as phishing emails and fake social media posts. Our data consisted of known scams such as phishing and other malicious activities. It came from FraudWatch Internationals Cyber Intelligence Datafeed, which collects information on cybercrime incidents.

Its tough to draw insights from unstructured data. Models cant easily discern between meaningful data points and those that are irrelevant or misleading (we call it noisy data). Data scientists rely on feature engineering to cut through the noise. This process identifies and labels meaningful data points using knowledge from other fields.

We used domain knowledge from behavioural science to engineer and label meaningful features in unstructured scam data. Scams were labelled based on how they used Cialdinis social influence principles, transforming raw text data into meaningful features. For example, a phishing email might use the principle of urgency by saying your account will be locked in 24 hours if you do not respond!. The raw text is transformed into a meaningful feature labelled urgency, which can be analysed for patterns. Then we used machine learning to analyse and visualise the labelled dataset.

The results showed that certain social influence principles such as liking and authority were frequently used together in scams. We also found that phishing scams often employed a mix of several principles. This made them more sophisticated and harder to detect.

The results gave us valuable insights into how often different types of social influence principles (such as urgency, trust, familiarity) are exploited by cybercriminals, as well as where more than one type is used at a time. Analysing unstructured text data like phishing emails and fake social media posts allowed us to identify patterns that indicated manipulative tactics.

Overall, our work yielded high quality insights from complex scam datasets.

Its important to mention that our dataset was not exhaustive. However, we believe our results are invaluable for mining insights from complex cybercrime data. This kind of analysis can be used by cybersecurity professionals, data scientists, cybersecurity firms and organisations involved in cybersecurity research. It can help improve automated detection systems and inform targeted training.

Visit link:

Catching online scammers: our model combines data and behavioural science to map the psychological games cybercriminals play - The Conversation Canada

11 min read

Are you a data scientist dreaming of landing a job in Big Tech but youre not sure what skills you need to get there?

Well, Ive got a secret weapon that could be just what you need to land your dream job in top tech companies.

A few months ago, I wrote this article about all the essential skills you need to get hired by the best tech firms, and today, were going to focus on one of those crucial skills: Experimentation.

Experimentation is a statistical approach that helps us isolate and evaluate the impact of product changes launching features, UX updates, and all!

But why is experimentation so important for standing out among other data scientists?

Its simple. The biggest tech companies are all about creating great products, and experimentation is a vital tool in achieving that.

If you can become an expert in experimentation, youll have a significant advantage over other candidates because most job seekers overlook this skill and dont know how to develop it.

Excerpt from:

Master This Data Science Skill and You Will Land a Job In Big Tech Part I | by Khouloud El Alami | Jul, 2024 - Towards Data Science

Consistent data processing: a critical prelude to building AI models

The critical nature of precise storage, management, and dissemination of data in the realm of drug discovery is universally recognized. This is because the extraction of meaningful insights depends on the data being readily accessible, standardized, and maintained with the highest possible consistency. However, the implementation of good data practices can vary greatly and depends on the goals, culture, resources, and expertise of research organizations. A critical, yet sometimes underestimated, aspect is the initial engineering task of data preprocessing, which entails transforming raw assay data into a format suitable for downstream analysis. For instance, quantifying sequencing reads from DNA-encoded library screens into counts is required for the subsequent hit identification data science analysis step. Ensuring the correctness of this initial data processing step is imperative, but it may be given too little priority, potentially leading to inaccuracies in subsequent analyses. Standardization of raw data processing is an important step to enable subsequent machine learning studies of DEL data. Currently, this step is done by companies or organizations that generate and screen DEL libraries, and the respective protocols are reported if a study is published (see the Methods section in McCloskey et al. 18). Making data processing pipelines open source will help establish best practices to allow for scrutiny and revisions if necessary. While this foundational step is vital for harnessing data science, it is worth noting that it will not be the focus of this discussion.

In drug discovery, data science presents numerous opportunities to increase the efficiency and speed of the discovery process. Initially, data science facilitates the analysis of huge experimental data, e.g., allowing researchers to identify potential bioactive compounds in large screening data. Machine learning models can be trained on data from DEL or ASMS and, in turn, be used for hit expansion in extensive virtual screens. For example, a model trained to predict the read counts of a specific DEL screen can be used to identify molecules from other large compound libraries, which are likely to bind to the target protein under consideration18.

As the drug discovery process advances to compound optimization, data science can be used to analyse and predict the pharmacokinetic and dynamic properties of potential drug candidates. This includes model-based evaluation of absorption, distribution, metabolism, excretion, and toxicity (ADMET) profiles. ADMET parameters are crucial in prioritizing and optimizing candidate molecules. Acknowledging their importance, the pharmaceutical industry has invested substantially in developing innovative assays and expanding testing capacities. Such initiatives have enabled the characterization of thousands of compounds through high-quality in-vitro ADMET assays, serving as a prime example of data curation in many pharmaceutical companies37. The knowledge derived from accumulated datasets has the potential to impact research beyond the projects where the data was originally produced. Computational teams utilize these data to understand the principles governing ADMET endpoints as well as to develop in-silico models for the prediction of ADMET properties. These models can help prioritize compound sets lacking undesired liabilities and thus guide researchers in their pursuit to identify the most promising novel drug candidates.

Major approaches in early drug discovery data science encompass classification, regression, or ranking models. They are, for example, employed in drug discovery to classify molecules as mutagenic, predict continuous outcomes such as the binding affinity to a target, and rank compounds in terms of their solubility. Incorporating prior domain knowledge can further enhance the predictive power of these models. Often, assays or endpoints that are correlated can be modelled together, even if they represent independent tasks. By doing so, the models can borrow statistical strength from each individual task, thereby improving overall performance compared to modelling them independently. For example, multitask learning models can predict multiple properties concurrently, as demonstrated by a multitask graph convolutional approach used for predicting physicochemical ADMET endpoints38.

When confronted with training data that have ambiguous labels, utilizing multiple-instance learning can be beneficial. Specifically, in the context of bioactivity models, this becomes relevant when multiple 3D conformations are considered, as the bioactive conformation is often unknown39. A prevalent challenge in applying data science for predictive modelling of chemical substances is choosing a suitable molecular representation. Different representations, such as Continuous Data-Driven Descriptor (CDDD)40 from SMILES strings, molecular fingerprints41 or 3D representations42, capture different facets of the molecular structure and properties43. It is vital to select an appropriate molecular representation as this determines how effectively the nuances of the chemical structures are captured. The choice of the molecular representation influences the prediction performance of various downstream tasks, making it a critical factor in AI-driven drug discovery, as discussed in detail in David et al.s43 review and practical guide on molecular representations in AI-driven drug discovery. Recent studies have found that simple k-nearest neighbours on molecular fingerprints can match or outperform much more complicated deep learning approaches on some compound potency prediction benchmarks44,45. On the other hand, McCloskey et al. 18 have discovered hits by training graph neural networks on data from DEL screens, which are not close to the training set using established molecular similarity calculations. Whether a simple molecular representation, infused with chemical knowledge, or a complex, data-driven deep learning representation is more suitable for the task at hand depends strongly on the training data and needs to be carefully evaluated on a case-by-case basis to obtain a fast and accurate model.

Sound strategies for splitting data into training and test sets are crucial to ensure robust model performance. These strategies include random splitting, which involves dividing the data into training and test sets at random, ensuring a diverse mix of data points in both sets. Temporal splitting arranges data chronologically, training the model on older data and testing it on more recent data, which is useful for predicting future trends. Compound cluster-wise splitting devides training and test sets into distinct chemical spaces. Employing these strategies is essential as inconsistencies between the distributions of training and test data can lead to unreliable model outputs, negatively impacting decision-making processes in drug discovery46.

The successful application of machine learning requires keeping their domain of applicability in mind at all stages. This includes using the techniques described in the previous section for data curation and model development. However, it is equally important to be able to estimate the reliability of a prediction made by an AI model. While generalization to unseen data is theoretically well understood for classic machine learning techniques, it is still an active area of research for deep learning. Neural networks can learn complex data representations through successive nonlinear transformations of the input. As a downside of this flexibility, these models are more sensitive to so-called adversarial examples, i.e., instances outside the domain of applicability that are seemingly close to the training data from the human perspective44. For this reason, deep learning models often fall short of providing reliable confidence estimates for their predictions. Several empirical techniques can be used to obtain uncertainty estimates: Neural network classifiers present a probability distribution indicative of prediction confidence, which is inadequately calibrated but can be adjusted on separate calibration data45. For regression tasks, techniques such as mixture density networks47 or Bayesian dropout48 can be employed to predict distributions instead of single-point estimates. For both classification and regression, the increased variance of a model ensemble indicates that the domain of applicability has been left49.

With the methods described in the previous paragraphs, we possess the necessary methodological stack to establish a data-driven feedback loop from experimental data, a crucial component for implementing active learning at scale. By leveraging predictive models that provide uncertainty estimates, we can create a dynamic and iterative data science process for the design-make-test-analyse (DMTA) cycle. For instance, these predictive models can be utilized to improve the potency of a compound by identifying and prioritizing molecules that are predicted to have high affinity yet are uncertain. Similarly, the models can be used to increase the solubility of a compound by selecting molecules that are likely to be more soluble, thus improving delivery and absorption. This process continuously refines predictions and prioritizes the most informative data points for subsequent experimental testing and retraining the predictive model, thereby enhancing the efficiency and effectiveness of drug discovery efforts. An important additional component is the strategy to pick molecules for subsequent experiments. By intelligently selecting the most informative samples, possibly those that the model is most uncertain about, the picking strategy ensures that each iteration contributes maximally to refining the model and improving predictions. For example, in the context of improving compound potency, the model might prioritize molecules that are predicted to have high potency but with a high degree of uncertainty. These strategies optimize the DMTA process by ensuring that each experimental cycle contributes to the refinement of the predictive model and the overall efficiency of the drug discovery process.

When applying the computational workflow depicted in Fig.3 on large compound libraries, scientists encounter a rather uncommon scenario for machine learning: usually, the training of deep neural networks incurs the highest computational cost since many iterations over large datasets are required, while comparatively few predictions will later be required from the trained model within a similar time frame. However, when inference is to be performed on a vast chemical space, we face the inverse situation. Assessing billions of molecules for their physicochemical parameters and bioactivity is an extremely costly procedure, potentially requiring thousands of graphics processing unit (GPU) hours. Therefore, not only predictive accuracy but also the computational cost of machine learning methods is an important aspect that should be considered when evaluating the practicality of a model.

Computational workflow for predicting molecular properties, starting with molecular structure encoding, followed by model selection and assessment, and concluding with the application of models to virtually screen libraries and rank these molecules for potential experimental validation. The process can be cyclical, allowing iterative refinement of models based on empirical data. ADMET: absorption, distribution, metabolism, and excretiontoxicity. ECFP: Extended Connectivity Fingerprints. CDDD: Continuous Data-Driven Descriptor, a type of molecular representation derived from SMILES strings. Entropy: Shannon entropy descriptors50,51.

Originally posted here:

A data science roadmap for open science organizations engaged in early-stage drug discovery - Nature.com