ABOUT THIS EPISODE

English
United States

TRANSCRIPT 🔗

SEARCH PAST EPISODES

Search past episodes of Artists of Intelligence with Harpreet Sahota.

OTHER EPISODES IN THIS PODCAST

Overview Machine learning models deployed in web applications operate with generous latency budgets, elastic cloud compute, and forgiving user expectations. Physical robotics operates in an entirely different reality: payload weight limits, strict thermal dissipation caps, intermittent network con…
Overview Generating music with deep learning exposes both the surprising effectiveness and the structural limits of autoregressive sequence modeling. When musical events are mapped into discrete tokens, transformers can learn harmony, polyphony, and instrumental timbre with remarkable fidelity. Ye…
Overview Graph Neural Networks (GNNs) solve a problem classical convolutional and transformer architectures cannot: modeling non-Euclidean, irregular relational topologies across nodes and edges. However, taking GNNs from academic benchmarks to high-throughput production environments introduces se…
Overview The transition from discriminative machine learning to generative foundation models broke conventional MLOps assumptions. In classical predictive modeling, 80% of engineering bandwidth was dedicated to upstream data wrangling, feature stores, and deterministic ground-truth evaluation. Wit…
Overview Real-time computer vision is fundamentally an optimization problem defined by hardware trade-offs. While research benchmarks celebrate minor improvements in mean Average Precision (mAP), production deployments demand low latency, minimal parameter footprints, and predictable export paths …
Disclaimer: The podcast and artwork embedded on this page are from Harpreet Sahota, which is the property of its owner and not affiliated with or endorsed by Listen Notes, Inc.