Arkadeep Banerjee

Staff Data Scientist |
Hobby Perfumer |
NLP, DL, RL, XAI, Private AI, Knowledge Graph |
Jadavpur University|
MTech(CS), ISI Kolkata

[Arkadeep Banerjee @arkadeep_thinks]

The resilience shown by the doctors in face of all adversities is extremely praiseworthy...

I have long held the belief that amongst all sections of the society, doctors, are probably the only ones who can effectualize changes for the betterment of common mass.. Lead the way 🙏

[Dr. Abhinaba Pal @drabhinabapal7]

I have never felt so helpless & frustrated! Look how my seniors are drenched in the rain. All they want is a transparent,fair, open discussion. The entire State is looking forward to this meeting but our CM #MamataBanerjee is creating all the mess!
#RGKarProtest #JusticeForRGKar

[Arkadeep Banerjee @arkadeep_thinks]

@blueinwe @UKafrolista Zoologist Bee, Wesker Eau de Mystique

[Arkadeep Banerjee @arkadeep_thinks]

���🤓Ask us Anything - Data Science / Machine Learning“ with @bhavishyap9, @karthiksr, Ishwar Bagga, @sahatamoghna, @mananvermatech, @arkadeep_thinks, Akriti Ahuja, Ankit Jha, Arvind Ramachandran, @apurv_saha, Vivek Jain. Today at 10:00 pm on @clubhouse! https://t.co/iGWY9YcdVA

[nasscom events @NasscomEvents]

Arkadeep Banerjee of Intuit talks a form of collaborative learning that can address the challenges of data inaccessibility and data privacy in the blog 'Federated Learning: A privacy-first approach to ML'.

Read the full article here: https://t.co/Za9iTHMCw7

[Arkadeep Banerjee @arkadeep_thinks]

@doesdatmaksense Keep up the great work Antaripa.. This is indeed flex worthy💪

[jeffheaton @jeffheaton]

Here is the latest PDF version of the free 500+ page textbook that I developed for my Applications of Deep Learning course that is about to start for Fall 2021 at @WashUengineers . #python #deeplearning #keras https://t.co/0ZQ27vTx7O

[Arkadeep Banerjee @arkadeep_thinks]

It will be interesting to see how the world of competitive programming evolves, when someone can create a browser plug-in that captures the screen, deciphers the task and calls the Codex API..

[OpenAI @OpenAI]

OpenAI Codex has a qualitatively new capability — it can write code with sufficient accuracy that users can direct their computer in natural language.

Codex is now in OpenAI API, for people to build new businesses or integrate with existing applications: https://t.co/n3U4DJ7fqC

[Arkadeep Banerjee @arkadeep_thinks]

One thing that should not exist in 2021 is *paid* online resources for learning...

For #DataScience enthusiasts, follow the below thread and the comments for free resources on various DS concepts and applications..

[Arkadeep Banerjee @arkadeep_thinks]

@bindureddy Abstraction of the Underlying Phenomenon

[Arkadeep Banerjee @arkadeep_thinks]

@SahaTamoghna @nandhaez @ash_ishh_ @BhavishyaP9 🙌🏽

[clubhousestats @clubhousestats]

@finny_vc @tek_fin @i_vp @jakefeinbaum @ExecutiveDre @solonang @ddurai7 @rohitsrivastv Top 10 #SaaS Clubs on #Clubhouse 🎉

6. Nuio Forum… - 4.3k 🎙
7. Tech Partnership Alliance - 3.8k 🥥
8. Data Science @ India - 3.4k 🇮🇳
9. Women In Sales - 3.2k 👩‍💼
10. Digital Business - 3.1k 🌱

@nuioHQ @B_BaderH @arkadeep_thinks @ash_ishh_ @mabdelgaber @sadekfarrag

[Arkadeep Banerjee @arkadeep_thinks]

@LucileSaulnier Thank you Saulnier 😊

[Arkadeep Banerjee @arkadeep_thinks]

@LucileSaulnier Thank you, this helps. One more question, how were you shuttling the data and gradient around? Was it completely taken care of by the hivemind implementation? or did you have some RPC based protocol in place?

[Arkadeep Banerjee @arkadeep_thinks]

@LucileSaulnier or Gossip Learning?

[Arkadeep Banerjee @arkadeep_thinks]

@LucileSaulnier How would one differentiate this from Federated Learning of LM, where one of the participant is the server (the aggregator) itself? In other words, will you please highlight the difference between FL and DeDLOC?

[Arkadeep Banerjee @arkadeep_thinks]

@almost_kapil @chipro True... I am not writing comments anymore 😜