I will be starting this exercise of annotating papers/Videos/Blogs to keep of with the latest industry research of Deep Learning, Computer Vision, and Machine Learning Systems, Search & Ranking, Information Retrieval, and Recommendation Systems, Personalization:

A. Deep Learning:

  • Embedding Representation
    - Contrastive Learning
    - Metric Learning
    - Multitask Learning

a) https://medium.com/pinterest-engineering/multi-task-learning-for-related-products-recommendations-at-pinterest-62684f631c12
b) https://medium.com/pinterest-engineering/how-we-use-automl-multi-task-learning-and-multi-tower-models-for-pinterest-ads-db966c3dc99e
c) https://labs.pinterest.com/user/themes/pin_labs/assets/paper/pintext-kdd2019.pdf

B. Computer Vision:

  • Visual Search

a) https://ai.facebook.com/blog/advancing-ai-to-make-shopping-easier-for-everyone/

b) https://engineering.fb.com/2018/05/02/ml-applications/advancing-state-of-the-art-image-recognition-with-deep-learning-on-hashtags/

  • Object detection and Segmentation

C. Machine Learning systems/Machine Learning Operations:

D. Search & Ranking, Information Retrieval, and Recommendation Systems, Personalization:

E. Mathematics of Machine Learning:

You can find those papers on this Github.

Continuing the journey in CS329S from Stanford I would be highlighting a few points and challenges of the ML production chapter from the course which I feel is essential to realize when you want to bring research to production.

Machine Learning Research vs Production:

Machine learning research or academia focuses solely on the single objective to improve model performance and achieve State of the Art (SOTA) on benchmarked datasets. While this is not the case for industry and production which needs to cater to other objectives like computational capability, data, fairness, and interpretability.

I am currently researching Machine Learning System Design Patterns. This blog presents my journey in understanding these design patterns. The main agenda in this learning various design patterns for model serving, training, and operation in machine learning production.

The main topics have been referenced from ML-Design-Pattern and this and I shall just summarize my findings and expand more on them. The second agenda would be summarizing the topics of the book. I have additional references mentioned below that might be useful to read more about.

Reference: https://blog.bismart.com/hubfs/20190903-MachineLearning.jpg

The series can be found here:

  1. Serving Patterns
  2. QA Patterns
  3. Training Patterns
  4. Lifecycle Patterns

Reference Links:

  1. https://mercari.github.io/ml-system-design-pattern/
  2. https://github.com/mercari/ml-system-design-pattern
  3. https://medium.com/swlh/ml-design-pattern-1-transform-9e82ccbc3209
  4. https://www.oreilly.com/library/view/machine-learning-design/9781098115777/
  5. https://www.educative.io/courses/grokking-the-machine-learning-interview/JY4x4vAV8yD
  6. https://lakshmanok.medium.com/machine-learning-design-patterns-58e6ecb013d7
  7. https://www.manning.com/books/deep-learning-design-patterns

You can reach out to me on Linkedin and view my work on Github.

This is my research and study in Machine Learning Systems.

Having worked as a Software Development Engineer in India and then moving for my Masters to focus my studies in Machine Learning at New York University, Courant Institute of Mathematical Sciences, the concept of Machine Learning always fascinated me, and technologies like Siri, Cortana propelled me to learn how to build and scale such systems into production. I had studied various MOOCS (CS231n, AndrewNg’s Course), courses, and projects at NYU in Computer Vision, Deep Learning, Natural Language Processing, Reinforcement Learning.

Though all these helped me get a hold and keeping…

Stealthily beneath the moonlight,
escaped to a world,unbiased,
invisible with all might,
chastened by my identity,
I shelved it far behind.
Hysterical with comfort this land gave,
I strode across,head held high,
the mind without fear,Euphoria!

Deepened every notion,
as lingered clouds of doubt.
surfed every troubled tide,
Hope as my beacon of light.
Thousands of intellects, I approached,
enlightened,I quenched my thirsty soul,
every nerve danced anonymous tones.

Both mirthful and mystified,
unaware of the future,
I graced the odyssey of life,
penultimate to the steps of the heavenly abode .

Frantically searched I,anyone,
who fathomed my loneliness.
my heart wined,a desire,
if someone missed me back,
days and nights,
months passed into years,
the hype of detachment falsified,
wondered I,
Incognito-a faux pas or,
the needed quantum of solace!

Gazing through, a dark abyss,

mirroring a black hole,

the mind once boggling,

unescapable, lost into the convolutions,

itself synthesized,

Indifferent to mundane matters,

ponders, if the extreme infinities lie on circle rather than a line?

evaluates, the choices of atheist, agnostic and a theist,

broods, the existence of something rather than nothing,

Contemplating, if actually their is free will?

Vocalizing, the possibilities of death being an after life!

Questioning endlessly, the purpose of life, a drag?

Calming the inner rage, to find a sense,

Clearing the chaos, suppressing the enlightened thoughts,

Away from quadrants of space and time,

The consciousness drifting away,

recursively, a mess of unconscious complexity ,

A convolution to keep, away from a convolution!

Reference: https://images.app.goo.gl/naSKw3DaKeZwsWR67

Each day as a growing teen,

I realized the world is mean.

Every path I traveled diverged,

“seek the end”,my mind urged.

For pain was my eternal companion,

we fought along,all through the canyon.

Oh! I was lost in the labyrinth,

Every thought pushed me deeper into it!

Solitude had engulfed me,

and cursed my perpetual decree.

Deafened by the sound of silence,

it reminded me of my defiance.

In pursuit of happiness,

tears eased my heart’s heaviness.

Oh! I was lost in the labyrinth,

How I wish someone tugged my heart string!

Broken,bruised and distraught,

i gazed…

The graduate record examination is one of the most frequented examination for the graduate schools. I had given it this july(2016). I would like to share my experience of how I scored 320 (q-165,v-155) and awa -3.5 all within a span of 35 days of studies. Introducing to my background of having a bachelor’s in computer science from Mumbai university, I had a strong hold on the quantitative section.The verbal section of GREthough being tough due to its complexity, it was never my strong point. …

This is one of the most important issue which everyone of us has experienced it or felt an inner voice that needed to be spoken out. Twenty two years of human existence and experiences as a school kid, college goer ,university student and corporate employee and being on both sides of the coin, I strongly resonate with the angry feelings of people loosing out to biasness. …

Sultry noon,claimed the month of June,

hastily centered in the hamlet,

rhythmically rung the bells of the temple,

with panache ,chanted the devotees along.

Sat on the deserted bench,a man,

his hair silvery white,spoke wise,

his head bowed down,

Tepidly I strolled on,

Eyes that never met,

A glimpse I could all get,

Struck! I witnessed him shed silent tears,

What grief had engulfed him,

despair? loneliness? A tragedy?

I could never know,

sympathy could I ever show?

I questioned to myself,

those tears,treasured by none?

ingratitude vehemently gossiped,

suffocated died the love,

love that knew no bounds?

indebted were those who indulged

in selfishness and greed,

How could you even forget,

the hands that feed?

could time ever repay time?

saddened,I penned down this,

an ode to that teary old man!

Sree Gowri Addepalli

Tech(Software, Systems, AI). Philosophy.Literature

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store