When normal people think of machine learning, they imagine the terminator or the matrix but when computer geek thinks of ai , they imagine big gpu clusters and tpus ,but that is not helpful and reliable for our home or iot devices as it has a lot of latency and privacy concerns ,that’s where TinyML comes in to save ML.
TinyML is mainly running ml on tiny devices which have power consumption less than milli watt and are super super cheap . No one needs to plug them to power source or check daily for maintenance as they are our tiny little helpers. They can help us solve problems with ml which would have been impossible due to normal ML like monitoring marine life density or behaviors of migrating animals or even help us detect patterns in our own health.
Think of wearing a single device which keeps the data local and detects anomaly in your daily bodily functions. I think this could be really helpful for detecting diseases before symptoms . I haven't talked really about the usefulness of it in our iot and home devices which I will leave it to you the reader.
Most of the stuff that requires machine learning(user based) could be done locally on the device with future advancements in our chipsets which would make our experience even more smooth.
One of the most common use case of TinyML is detecting anomaly in the functioning of machines for preventing machine breakdown and waking up when you say “Hey Siri” or “Hey Google”.
Machine learning is about empowering us to solve problems that we could only think of earlier and TinyML only increases that further.
I would like to Thank Pete Warden for reviewing the blog and inspiring me with his book about TinyML .
And I have been chosen as one of the Ambassadors of Full Steam Ahead so if you have an existing STEM team Signup at https://forms.gle/Gr31PUz5koNfjaCZA
When normal people think of machine learning, they imagine the terminator or the matrix but when computer geek thinks of ai , they imagine big gpu clusters and tpus ,but that is not helpful and reliable for our home or iot devices as it has a lot of latency and privacy concerns ,that’s where TinyML comes in to save ML.
TinyML is mainly running ml on tiny devices which have power consumption less than milli watt and are super super cheap . No one needs to plug them to power source or check daily for maintenance as they are our tiny little helpers. They can help us solve problems with ml which would have been impossible due to normal ML like monitoring marine life density or behaviors of migrating animals or even help us detect patterns in our own health.
Think of wearing a single device which keeps the data local and detects anomaly in your daily bodily functions. I think this could be really helpful for detecting diseases before symptoms . I haven't talked really about the usefulness of it in our iot and home devices which I will leave it to you the reader.
Most of the stuff that requires machine learning(user based) could be done locally on the device with future advancements in our chipsets which would make our experience even more smooth.
One of the most common use case of TinyML is detecting anomaly in the functioning of machines for preventing machine breakdown and waking up when you say “Hey Siri” or “Hey Google”.
Machine learning is about empowering us to solve problems that we could only think of earlier and TinyML only increases that further.
I would like to Thank Pete Warden for reviewing the blog and inspiring me with his book about TinyML .
And I have been chosen as one of the Ambassadors of Full Steam Ahead so if you have an existing STEM team Signup at https://forms.gle/Gr31PUz5koNfjaCZA
Madhav GoyalHalo: A privacy layer for stablecoins
AbstractHalo is a privacy layer for stablecoins unlocking the next private neobanks and applications that were blocked by the lack of anonymity and confidentiality of networks. Halo Network introduces a new extension a a privacy-preserving EVM chain purpose-built for stablecoin use cases. Halo combines the familiarity and composability of the EVM with the confidentiality of a UTXO-based privacy layer, allowing users to transact and build privately without sacrificing interoperability or liqui...
Madhav Goyal🧊Solana::Leader schedule
I have been diving into Solana and its Validator code realizing the incredible design behind the selection of Validators and block producers, so here is a small rundownWhat are Blocks?What are POW and POS?Solana Leader ScheduleSpecial Considerations to produce the Tastiest BlockOpen QuestionsSome ResourcesBlocks wat?Blocks containing transactions and state transition data lined on top of one another with cryptography securing their computational integrity is the foundation of blockchain aka c...
Madhav GoyalMusings on Price Discovery
Musings on Price discoveryWith a fading bull run comes the pressure to do a token generation event at high valuations because otherwise if you are a VC backed project with 8 figs in funding, you are cooked without a binance listing since you definitely know your tech is only valued at the number of CEX listings and shady market makers you can get for the launch. Jokes and criticism aside, I have been diving into the price discovery for a mix of work and my own interests in the past few months...
Madhav GoyalHalo: A privacy layer for stablecoins
AbstractHalo is a privacy layer for stablecoins unlocking the next private neobanks and applications that were blocked by the lack of anonymity and confidentiality of networks. Halo Network introduces a new extension a a privacy-preserving EVM chain purpose-built for stablecoin use cases. Halo combines the familiarity and composability of the EVM with the confidentiality of a UTXO-based privacy layer, allowing users to transact and build privately without sacrificing interoperability or liqui...
Madhav Goyal🧊Solana::Leader schedule
I have been diving into Solana and its Validator code realizing the incredible design behind the selection of Validators and block producers, so here is a small rundownWhat are Blocks?What are POW and POS?Solana Leader ScheduleSpecial Considerations to produce the Tastiest BlockOpen QuestionsSome ResourcesBlocks wat?Blocks containing transactions and state transition data lined on top of one another with cryptography securing their computational integrity is the foundation of blockchain aka c...
Madhav GoyalMusings on Price Discovery
Musings on Price discoveryWith a fading bull run comes the pressure to do a token generation event at high valuations because otherwise if you are a VC backed project with 8 figs in funding, you are cooked without a binance listing since you definitely know your tech is only valued at the number of CEX listings and shady market makers you can get for the launch. Jokes and criticism aside, I have been diving into the price discovery for a mix of work and my own interests in the past few months...
Just a tinkerer in this wonderful world
Just a tinkerer in this wonderful world
Share Dialog
Share Dialog
Subscribe to Madhav Goyal
Subscribe to Madhav Goyal
<100 subscribers
<100 subscribers
No activity yet