LATEST POSTS

Model quantization is a technique used to optimize machine learning models by reducing their size and computational requirements, making them more efficient and suitable for deployment on resource-constrained devices such as smartphones, edge devices, and embedded systems. Quantization involves converting the model’s parameters (weights and activations) from high-precision data types, such as 32-bit...

ESP32 is a single 2.4 GHz Wi-Fi-and-Bluetooth combo chip designed with the TSMC low-power 40 nm technology. We will use the GPIO pins functionality only in this post, but each pin has more than one possible functionality and while using a pin for particular task, we need to double check its alternative functions. Most of the GPIO pins have PWN functionality except pins (34,35,36,39), some GPIO pins...

The physical design process is crucial in transforming chip designs into real-world products that drive our digital world. It involves a carefully planned series of steps aimed at optimizing chip layouts for performance, power efficiency, and manufacturability. But what are these steps, and how do they fit into the broader design process? Let’s dive into the complexities of physical design and explore...

A cryptocurrency is a digital currency designed to work as a medium of exchange through a computer network that is not reliant on any central authority, such as a government or bank, to uphold or maintain it. It has, in a financial point of view, grown to be its own asset class. However, on the contrary to other asset classes like equities or commodities, sectors have not been officially defined as...

A generative adversarial network (GAN) is a class of machine learning frameworks and a prominent framework for approaching generative AI.The core idea of a GAN is based on the “indirect” training through the discriminator, another neural network that can tell how “realistic” the input seems, which itself is also being updated dynamically. This means that the generator is not...

When the transformer model came in the “Attention is all you need” paper, it changed the way NLP tasks were handled. I explained it in my previous post Transformer modelIn 2021 there was another paper “An image is worth 16×16 words” to attempt doing computer vision tasks using the transformer model. The model ignores the traditional approach of filters used for CNNs and...

A neural network is a network of neurons. A neuron is a mathematical function which transforms the input data elements into a single output value. In the above diagram the single neuron applies a given weight-age and a bias to each input to come up with the output value. If we represent the neuron bias as another input to neuron then the above equation can be simplified as since b1 is another weight...

Public-key cryptography, or asymmetric cryptography, is the field of cryptographic systems that use pairs of related keys. Each key pair consists of a public key and a corresponding private key. Key pairs are generated with cryptographic algorithms based on mathematical problems termed one-way/trap-door functions. Security of public-key cryptography depends on keeping the private key secret, the public...

A Symmetric cipher is an algorithm which uses only 1 key for both encryption and decryption. So the same key is used to convert the plain text to encoded text and to convert the encoded text back to plain text.This is under assumption that the key remains in safe hands, the key is easy to remember or convey in a secure manner.The algorithms which use the keys are generally known, or can be acquired...

Let’s simplify how webRTC works. There are 3 parts to webRTC. Signalling Connecting Communicating WebRTC is a peer to peer protocol which eliminates the need for the traffic to be routed through a central server. The server is used only to help setup the initial channels. Its called a signalling server since it builds up the initial handshakes. When 2 users want to communicate using webRTC, they...