ODSC East 2022
April 19th-21st
More sessions coming soon
East 2022 Preliminary Schedule
We are delighted to announce our East 2022 Schedule!
Training | In-Person | AI for Engineers | Machine Learning | Intermediate-Advanced
The lifecycle of any machine learning model, regular or deep, consists of (a) the pre-processing/transformation/augmenting of data (b) the training of the model with different hyper-parameter values/learning rates (c) the computing of results on new data/test sets. Whether you are using transfer learning, or a from-scratch model, this process requires a large amount of computation, management of your experimental process, and the quick perusal of results from your experiment. In this workshop, we will learn how to combine off-the-shelf clustering software such as kubernetes and dask, with learning systems such as tensorflow/pytorch/scikit-learn, on cloud infrastructure such as AWS/Google Cloud/Azure to construct a machine-learning system for your data science team. We’ll start with an understanding of kubernetes, move on to analysis pipelines in sklearn and dask, finally arrive at kubeflow. Participants should install minikube on their laptops (https://kubernetes.io/docs/tasks/tools/install-minikube/), and create accounts on the Google Cloud…more details
Bootcamp | Virtual |Deep Learning | All Levels
The emergence of data science as a discipline has impacted businesses in a range of different ways. One primary impact has been to elevate the use of data in decision-making by using statistical methods to assess the ever-growing datasets companies are collecting. This workshop will review and introduce statistical techniques and touch on more advanced methods for dealing with noisy data and applying real-world constraints to analyses. This workshop assumes a working knowledge of standard statistical methods and will aim to connect theory to practice using real-world examples.
Lesson 1: Descriptive statistics and exploring data statistically
– (Re)familiarize yourself with basic descriptive statistics
– Use simple data exploration techniques to identify problems and limitations of a new dataset
Lesson 2: Statistical analyses...more details
Tutorial | Virtual | NLP | Beginner – Intermediate
This talk aims to give an overview walkthrough of the suite of NLP methods grounded in neural-network architectures, including recurrent neural networks (RNNs), transformers, and convolutional neural networks (CNNs). We will connect them by diving into their similarities and differences. You will come away from the talk gaining the overview picture of NLP and grasping the theoretical essence that underpins NLP methods. This talk hopes to empower you with the foundational NLP knowledge and reduce the knowledge barrier for you to jumpstart your NLP projects.
Training | In-Person | Data Visualization | Machine Learning | Intermediate
Upon completing this tutorial, you will be:
– familiar with how to use the NetworkX and nxviz Python packages for modelling and rationally visualizing networks,
– able to load node and edge data from a Pandas dataframe,
– familiar with object-oriented and matrix-oriented representations of graphs,
able to find paths between nodes, interesting structures in graphs, and projections of bipartite graphs.
– (if time permits) able to use matrix operations to simulate diffusion of information on networks..more details
Workshop | In-Person | Machine Learning | Intermediate
This workshop will feature a comprehensive tutorial on using the CatBoost library. We will walk you through all the steps of building a good predictive model…more details
Training | In-Person | Machine Learning Safety and Security | Intermediate
In this workshop, which is directed to both a Data Science audience who may want to learn DFIR, and a DFIR audience who may want to learn Data Science, Jess Garcia will explain the fundamentals of Data Science and DFIR, and will lead the audience through all the different steps of an end-to-end investigation using exclusively Data Science tools and techniques. In the process, Jess will introduce multiple forensic artifacts and will explain the value they provide to the overall investigation….more details
Workshop | In-Person | MLOps & Data Engineering | Beginner-Intermediate
Although powerful, modern machine learning models can be sensitive. Seemingly subtle changes in a data distribution can destroy the performance of otherwise state-of-the art models, which can be especially problematic when ML models are deployed in production.. In this workshop, we will give a hands-on overview to drift detection, the discipline focused on detecting such changes. We will start by building an understanding of the ways in which drift can occur, and why it pays to detect it. We’ll then explore the anatomy of a drift detector, and learn how they can be used to detect drift in a principled manner.
Half-Day Training | Virtual | MLOps & Data Engineering | Intermediate
Apply machine learning models to different engineering areas has been of particular interests for many data science practitioners and software engineers. As one of the mostly popular machine learning framework, TensorFlow and Keras have been widely used in many production environments for its robustness and scalability. In this session, we will provide a tutorial on TensorFlow and Keras, and guide your through a series of hands-on examples ranging from basic MNIST dataset to time series processing for model building. We will also cover data input and output processing with TensorFlow, from processing simple CSV files to cloud data warehouse services such as Google Cloud BigQuery. As a bonus we will also cover the integration of TensorFlow with Apache Kafka, to illustrate the streaming data pipeline that is used broadly across the industry.
Tutorial | In-Person | Data Analytics | Beginner – Intermediate
In this tutorial session, attendees will learn how a set of open source tools can be leveraged to perform standardization, characterization, and data quality assessment for various health data sources. Open source tools including Synthea, ETL-Synthea, Achilles, Data Quality Dashboard, and Ares will be reviewed and demonstrated in a data operations pipeline. We will demonstrate how the global health information community leverages this strategy to ensure research-ready health data.
Workshop | In-Person | Data Visualisation | Intermediate-Advanced
Data visualization is a powerful tool for facilitating confident, informed decision-making. ggplot2 is one of the most popular data visualization packages in use today. Based on comprehensive grammar and syntax, ggplot2 gives you the ability to create data visualizations quickly and iteratively, whether it’s a simple bar-chart or a complicated network analysis. This workshop will teach you how to manipulate and structure your data for visualizations, graph elements, and their associated terminology, how to select the appropriate graph based on your data, and how to avoid common graphing mistakes. You will also learn how to customize data visualizations and give them the ‘personal touches’ that make them memorable to your audience.
Workshop | Virtual | Machine Learning | Intermediate
Today data scientists use a rapidly evolving and diverse set of tools and platforms to build advanced analytic and machine learning models. After the model building and development, these models need to be available to business users to draw insights and make better decisions to generate value for the business. As a result, enabling business users to interact with these sophisticated models to create their own on-demand analytics is extremely critical to reducing time to value for any business. During this workshop, we demonstrate how data scientists can leverage Tableau analytics extensions API as a model-agnostic deployment platform to enable business users to interact dynamically with the bespoke ML models and to build their own on-demand analytics in a self-service manner.
Bootcamp | Virtual | Open-source | Beginner
By completing this workshop, you will develop an understanding of relational models of data, how SQL is used to retrieve that data, and how to join tables, aggregate information, and answer data science questions. You will also become familiar with many of the common types of SQL databases, how to access information in a database from the command line, and how to integrate database access from within Python…more details
Training | In-Person | Machine Learning | Beginner-Intermediate
The rapid progress in machine learning (ML) and the massive increase in the availability and diversity of data has enabled novel approaches to quantitative investment. It has also increased the demand for the application of data science to develop both discretionary and algorithmic trading strategies.
In this workshop, we will cover popular use cases for ML in the investment industry, and how data science and ML fit into the workflow of developing a trading and investment strategy from the identification and combination of alpha factors to strategy backtesting and asset allocation.
The workshop uses Python and various standard data science and machine learning libraries like pandas, scikit-learn, gensim, spaCy as well as TensorFlow and Keras. The code examples will be presented using jupyter notebooks and are based on my book ‘Machine Learning for Algorithmic Trading’…more details
Workshop | In-Person | Deep Learning | Intermediate – Advanced
The speed at which intelligent decision-making can take place is changing the very fabric of modern industry. From autonomous routing decisions for real-time traffic avoidance to intelligent integrated systems for customer service and assisted analysis or forecasting, the time to correct decisions can be a huge differentiator. During the course of this workshop, you will learn how to harness the power of Apache Spark to build up a data ingestion pipeline that can process and learn from streaming data. These learnings can then be applied to make simple decisions using SparkSQL in a parallel streaming system…more details
Tutorial | In-Person | Machine Learning | Responsible AI | Beginner-Intermediate
Bias is everywhere – in data, in algorithms, in humans. Therefore dealing with it is a difficult issue and ever more important now that data increases exponentially in nearly all dimensions (velocity, volume, veracity) and machine learning systems trained on these data sets are becoming omnipresent. It is however not as straightforward to be accomplished as you have to start with awareness, take care of data quality while being proficient in measuring and monitoring your algorithms…more details
Tutorial | In-Person| Beginner-Intermediate
In this tutorial, we will look at some of the recent advances in multi-task RL. We will start by discussing the nuances of the relationship between single-task and multi-task setup. For example, a multi-task RL problem can be modeled as a single-task RL problem. In that setup, all the multi-task environments are considered different parts of a large single task environment. We will see how this observation can be used to develop some general components and strategies that can convert any single task RL algorithm into a multi-task agent…more details
Training | Virtual | Data Visualization | Intermediate-Advanced
In this workshop, we will build an interactive data visualization from scratch using d3.js in the browser. The possibilities shown in d3 examples are exciting but the API surface of d3 and the various browser standards like HTML, CSS, SVG, and JavaScript, can be overwhelming. Think of this workshop as a guided tour that will point out the important things to pay attention to as we go step-by-step from CSV file to interactive visualization....more details
Training | In-Person | Intermediate-Advanced
Spark NLP is a Natural Language Processing (NLP) library built on top of Apache Spark ML. It provides simple, performant & accurate NLP annotations for machine learning pipelines that can scale easily in a distributed environment. Spark NLP comes with 3000+ pretrained pipelines and models in more than 200+ languages. It supports nearly all the NLP tasks and modules that can be used seamlessly in a cluster. Downloaded more than 1 million every month and experiencing 20x growth for the last one year, Spark NLP is used by 54% of healthcare organizations as the world’s most widely used NLP library in the enterprise. In this talk, Veysel will conduct a hands-on session to go over the library’s healthcare components and teach how to solve any NLP problem in healthcare with state-of-the-art methods and practices across the industry. He will also explain the best practices for building production-grade solutions around the latest research…more details
Talk | Virtual | MLOps & Data Engineering | Beginner – Intermediate
Surrounding a typical ML pipeline many details are commonly swept under the rug. How will we monitor production data for concept drift? How do we measure false negative rate in production? How confident can we be of our performance assessments with a small test set and how should they be modified when faced with biased data? How can we ensure our model follows reasonable assumptions? We introduce a new general purpose tool, the Model Validation Toolkit, for common tasks involved in model validation, interpretability, and monitoring. Our utility has submodules and accompanying tutorials on measuring concept drift, assigning and updating optimal thresholds, determining credibility of performance metrics, compensating for data bias, and performing sensitivity analysis. In this session, we will give a tour of the framework’s core functionality and some associated use cases.
Business Talk | Virtual | Cross Industry | Beginner
In a talk that guides the audience through the key strategies, strengths and challenges around NLP, Dr. Sameer Maskey will deliver the following key takeaways: Impact and application of NLP by enterprise across various industries Key NLP techniques that are being leveraged to accelerate business success The do’s and don’ts of adopting and deploying NLP techniques and how it can take businesses to the next level.
Talk | In-Person | Intermediate
Dask is the leading Python-native framework for distributed computing with a growing open source community. Dask is used by commercial enterprises and scientific communities for scheduling and coordinating task execution for data engineering and data science pipelines…more details
Talk | Virtual | Cross-Industry | Beginner-Intermediate
This talk will draw on information from the following two frameworks – IEEE 7000-2021: IEEE Standard Model Process for Addressing Ethical Concerns during System Design, and Section A of the Technical Best Practices from the Foundation for Best Practices in Machine Learning. In melding the more technically focused IEEE framework with the more business-oriented section from the Technical Best Practices framework, this talk aims to provide participants with practical and actionable guidance on designing effective and ethical AI products. The talk will focus on the initial stage of the AI product lifecycle. Specifically, we’ll focus on the following areas of the AI design process: team composition and roles, problem statement and solution mapping, integrating context, organizational capacity, and defining the product and outcomes. Participants will leave this session with a broad understanding of two prevailing ethical design frameworks and a practical understanding of the initial steps organizations can take to design ethical and effective AI systems.
Business Talk | Virtual | Cross-Industry | Beginner-Intermediate
This 30-minute business and strategy-oriented talk aims to educate data scientists and analytics leaders about the role and benefits of human-in-the-loop (HITL), presented from the standpoint of a 25+ year veteran of traditional machine learning applications who has recently been immersed in this new world. The session will include brief examples in application areas as diverse as Medical AI, Ag Tech, and autonomous vehicles.
Talk | In-Person | Deep Learning | Intermediate
Although initially a slow adopter of machine learning and computer vision, agriculture has become an important domain for these approaches. Computer vision is now a key element of agricultural systems to determine crop type, count plants, guide harvesting robots, identify issues like crop stress and weeds, and forecast yield. Adoption and extension of these approaches is critical due to the challenges facing global agriculture: the world’s population is predicted to reach 9.7billion by 2050, water supply is expected to fall 40% short of global needs by 2030, and climate change produces significant challenges and uncertainty. Fortunately advances in deep learning and remote sensing technologies have unlocked unprecedented opportunities for precision agriculture. However, challenges remain in leveraging common SOTA approaches, which are often developed for natural scene imagery like Imagenet, with remote sensing data; this data may be massive, spatiotemporal, multispectral, contain very small objects, and possess fundamentally different statistics than natural scene images. In this session we’ll explore some of these challenges around remote sensing data for precision agriculture and approaches for addressing them including spatiotemporal modeling, self-supervised and contrastive learning, and multi-task learning within a deep learning framework.
Talk | Virtual | Deep Learning | All Levels
Most evaluation suites for multi-agent reinforcement learning do not assess generalization to novel situations as their primary objective (unlike supervised-learning benchmarks). The subject of this talk, Melting Pot, is an evaluation suite that fills this gap, and is scalable because it uses reinforcement learning to reduce the human labor required to create novel test scenarios. This works because one agent’s behavior constitutes (part of) another agent’s environment. Melting Pot currently contains 85 unique test scenarios covering a broad range of topics such as social dilemmas, reciprocity, resource sharing, and task partitioning. Melting Pot is open source and free to use for anyone.
Talk | In-Person | Machine Learning | All Levels
Risk assessment is a popular task when machine learning is used for automated decision making. For example, Jack’s risk of defaulting on a loan is 8, Jill’s is 2; Ed’s risk of recidivism is 9, Peter’s is 1. We know that this task definition comes with impossibility results for group fairness, where one cannot simultaneously satisfy desirable probabilistic measures of fairness. I will highlight recent findings in terms of these impossibility results. Next, I will present work on how machine learning can be used to generate aspirational data (i.e., data that are free from biases of real-world data). Such data are useful for recognizing and detecting sources of unfairness in machine learning models besides biased data. Time-permitting, I will discuss steps in measuring our algorithmically infused societies.
Talk | Virtual | Responsible AI | Machine Learning Safety and Security | Intermediate
Under the umbrella of trustworthy computing, employing formal methods for ensuring trust properties such as reliability and security has led to scalable success. Just as for trustworthy computing, formal methods could be an effective approach for building trust in AI-based systems. However, we would need to extend the set of properties to include fairness, robustness, and interpretability, etc.; and to develop new verification techniques to handle new kinds of artifacts, e.g., data distributions and machine-learned models. This talk poses a new research agenda, from a formal methods perspective, for us to increase trust in AI systems.
Talk | In-Person | Machine Learning | Responsible AI | Intermediate
In this talk I will discuss how predictive analytics and machine learning can be utilized to identify patients appropriate for a palliative care path. I will then describe a particular use case – developing a palliative care referral mechanism that works well for both commercially insured adults and Medicare Advantage plan members – and use it to illustrate how data scientists can modify their modeling processes for success. I will discuss several approaches to modeling and demonstrate how each can be adjusted to work well within the bounds of multiple variations of one problem. I will also illustrate the limits of several common predictive algorithms in building niche models. Along the way I will give several deep dives into the practical challenges associated with predicting rare events such as palliative care need and discuss the limits of using claims data to develop measurable proxies for palliative care need and other health related events.