Access and Feeds

Blog Archives

Artificial Intelligence: Quality of Training Data is Key to Success

By Dick Weisinger

Artificial Intelligence (AI) projects create a model and then feed the model data. A general rule of thumb is that the more data that is input into the model, the better the results. But one thing noticed is that there…

Artificial Intelligence: Neural Network Short-Cuts Can Lead to Failures

By Dick Weisinger

Artificial Intelligence has made great progress over the last decade and is being used in applications like autonomous driving and medical diagnoses. But there has been a reluctance to trust or turn over control to AI algorithms because of a…

Explainable Artificial Intelligence: Can You Believe the Explanation? Are Validation and Verification a More Practical Approach?

By Dick Weisinger

Artificial Intelligence is trying to overcome the problem of explainability. AI algorithms have grown in complexity to include billions of tunable parameters, and have achieved results that are often considered stunning or amazing. But the problem is trust. For most…

Deep Learning: Researchers Pursue Advantages of Analog Computing

By Dick Weisinger

During the last decade Artificial Intelligence, particularly Deep Learning algorithms, have made enormous progress by adopting GPU processors. GPUs, graphic processing units, are just that — they were designed to boost the performance of computer graphics processing. Over the last…

AI Chips: AI-Targeted Massive Wafers Speed the Creation of Large Language Models

By Dick Weisinger

‘Wafer-scale engine‘ (WSE) – it is what Cerebras, a semiconductor manufacturer, calls the architecture of their massive GPU engine on a single ‘chip’. It is 56 times the size of the largest ‘standard’ GPU chip and packs 3000 times more…