Machine Learning

Build Better Machine Learning Models By Understanding Their Decisions With SHAP - Episode 335

Machine learning and deep learning techniques are powerful tools for a large and growing number of applications. Unfortunately, it is difficult or impossible to understand the reasons for the answers that they give to the questions they are asked. In order to help shine some light on what information is being used to provide the outputs to your machine learning models Scott Lundberg created the SHAP project. In this episode he explains how it can be used to provide insight into which features are most impactful when generating an output, and how that insight can be applied to make more useful and informed design choices. This is a fascinating and important subject and this episode is an excellent exploration of how to start addressing the challenge of explainability.

Read More

Accelerating Drug Discovery Using Machine Learning With TorchDrug - Episode 334

Finding new and effective treatments for disease is a complex and time consuming endeavor, requiring a high degree of domain knowledge and specialized equipment. Combining his expertise in machine learning and graph algorithms with is interest in drug discovery Jian Tang created the TorchDrug project to help reduce the amount of time needed to find new candidate molecules for testing. In this episode he explains how the project is being used by machine learning researchers and biochemists to collaborate on finding effective treatments for real-world diseases.

Read More

An Exploration Of Automated Speech Recognition - Episode 333

The overwhelming growth of smartphones, smart speakers, and spoken word content has corresponded with increasingly sophisticated machine learning models for recognizing speech content in audio data. Dylan Fox founded Assembly to provide access to the most advanced automated speech recognition models for developers to incorporate into their own products. In this episode he gives an overview of the current state of the art for automated speech recognition, the varying requirements for accuracy and speed of models depending on the context in which they are used, and what is required to build a special purpose model for your own ASR applications.

Read More

Experimenting With Reinforcement Learning Using MushroomRL - Episode 332

Reinforcement learning is a branch of machine learning and AI that has a lot of promise for applications that need to evolve with changes to their inputs. To support the research happening in the field, including applications for robotics, Carlo D’Eramo and Davide Tateo created MushroomRL. In this episode they share how they have designed the project to be easy to work with, so that students can use it in their study, as well as extensible so that it can be used by businesses and industry professionals. They also discuss the strengths of reinforcement learning, how to design problems that can leverage its capabilities, and how to get started with MushroomRL for your own work.

Read More

Monitor The Health Of Your Machine Learning Products In Production With Evidently - Episode 330

You’ve got a machine learning model trained and running in production, but that’s only half of the battle. Are you certain that it is still serving the predictions that you tested? Are the inputs within the range of tolerance that you designed? Monitoring machine learning products is an essential step of the story so that you know when it needs to be retrained against new data, or parameters need to be adjusted. In this episode Emeli Dral shares the work that she and her team at Evidently are doing to build an open source system for tracking and alerting on the health of your ML products in production. She discusses the ways that model drift can occur, the types of metrics that you need to track, and what to do when the health of your system is suffering. This is an important and complex aspect of the machine learning lifecycle, so give it a listen and then try out Evidently for your own projects.

Read More

Making Automated Machine Learning More Accessible With EvalML - Episode 329

Building a machine learning model is a process that requires a lot of iteration and trial and error. For certain classes of problem a large portion of the searching and tuning can be automated. This allows data scientists to focus their time on more complex or valuable projects, as well as opening the door for non-specialists to experiment with machine learning. Frustrated with some of the awkward or difficult to use tools for AutoML, Angela Lin and Jeremy Shih helped to create the EvalML framework. In this episode they share the use cases for automated machine learning, how they have designed the EvalML project to be approachable, and how you can use it for building and training your own models.

Read More

Delivering Deep Learning Powered Speech Recognition As A Service For Developers At AssemblyAI - Episode 326

Building a software-as-a-service (SaaS) business is a fairly well understood pattern at this point. When the core of the service is a set of machine learning products it introduces a whole new set of challenges. In this episode Dylan Fox shares his experience building Assembly AI as a reliable and affordable option for automatic speech recognition that caters to a developer audience. He discusses the machine learning development and deployment processes that his team relies on, the scalability and performance considerations that deep learning models introduce, and the user experience design that goes into building for a developer audience. This is a fascinating conversation about a unique cross-section of considerations and how Dylan and his team are building an impressive and useful service.

Read More

Exploring The SpeechBrain Toolkit For Speech Processing - Episode 323

With the rising availability of computation in everyday devices, there has been a corresponding increase in the appetite for voice as the primary interface. To accomodate this desire it is necessary for us to have high quality libraries for being able to process and generate audio data that can make sense of human speech. To facilitate research and industry applications for speech data Mirco Ravanelli and Peter Plantinga are building SpeechBrain. In this episode they explain how it works under the hood, the projects that they are using it for, and how you can get started with it today.

Read More

Lightening The Load For Deep Learning With Sparse Networks Using Neural Magic - Episode 321

Deep learning has largely taken over the research and applications of artificial intelligence, with some truly impressive results. The challenge that it presents is that for reasonable speed and performance it requires specialized hardware, generally in the form of a dedicated GPU (Graphics Processing Unit). This raises the cost of the infrastructure, adds deployment complexity, and drastically increases the energy requirements for training and serving of models. To address these challenges Nir Shavit combined his experiences in multi-core computing and brain science to co-found Neural Magic where he is leading the efforts to build a set of tools that prune dense neural networks to allow them to execute on commodity CPU hardware. In this episode he explains how sparsification of deep learning models works, the potential that it unlocks for making machine learning and specialized AI more accessible, and how you can start using it today.

Read More

Traversing The Challenges And Promise Of Graph Machine Learning - Episode 319

The foundation of every ML model is the data that it is trained on. In many cases you will be working with tabular or unstructured information, but there is a growing trend toward networked, or graph data sets. Benedek Rozemberczki has focused his research and career around graph machine learning applications. In this episode he discusses the common sources of networked data, the challenges of working with graph data in machine learning projects, and describes the libraries that he has created to help him in his work. If you are dealing with connected data then this interview will provide a wealth of context and resources to improve your projects.

Read More