Adversarial NLI: A New Benchmark for Natural Language Understanding

Nie et al.: https://arxiv.org/abs/1910.14599

#NaturalLanguageUnderstanding #MachineLearning #DeepLearning

Mechanical Engineer
https://aijobs.com/jobs/mechanical-engineer/

https://t.iss.one/ArtificialIntelligenceArticles

What does it mean for a machine to “understand”?
The use of words like “real”, “true”, and “genuine” imply that “understanding” is binary ... I argue that “understanding” exists along a continuous spectrum of capabilities.
In order for a system to understand, it must create linkages between different concepts, states, and actions. Today’s language translation systems correctly link “water” in English to “agua” in Spanish, but they don’t have any links between “water” and “electric shock”.
Seeking a definition of intelligence without including an agent’s environment is like seeking the sound of one hand clapping.
The other motivation of definitions of intelligence is that of Occam’s razor. This is what motivates the idea that compression is equal to intelligence.
Blogs
https://medium.com/@tdietterich/what-does-it-mean-for-a-machine-to-understand-555485f3ad40
https://medium.com/intuitionmachine/how-to-define-intelligence-b9bac630960b

Quantum supremacy using a programmable superconducting processor - J. Martinis - 11/1/2019


IQIM Seminar by John Martinis (Research Scientist and Professor of Physics Google and University of California, Santa Barbara), "Quantum supremacy using a programmable superconducting processor"

https://www.youtube.com/watch?v=FklMpRiTeTA

https://t.iss.one/ArtificialIntelligenceArticles

TensorFlow World 2019 Keynote
https://www.youtube.com/watch?v=MunFeX-0MD8&list=PLQY2H8rRoyvxcmHHRftsuiO1GyinVAwUg

Making an Invisibility Cloak for evading Object Detectors!

https://www.profillic.com/paper/arxiv:1910.14667

(eg.the YOLOv2 detector is evaded using a pattern trained on the COCO dataset with a carefully constructed objective.)

Btw if you're interested in implementing this in your project/product, feel free to DM me

Limitations of the Empirical Fisher Approximation

Kunstner et al.: https://arxiv.org/abs/1905.12558

#ArtificialIntelligence #MachineLearning

Submitted to WACV 2020: Turning low-resolution pictures to super high resolution

https://www.profillic.com/paper/arxiv:1910.08761

a fully convolutional multi-stage neural network for 4× super-resolution for face images.

On the Interaction Between Deep Detectors and Siamese Trackers in Video Surveillance
Kiran et al.: https://arxiv.org/abs/1910.14552
#ArtificialIntelligence #DeepLearning #MachineLearning

Deep Learning course: lecture slides and lab notebooks
Built and maintained by Olivier Grisel and Charles Ollion: https://m2dsupsdlclass.github.io/lectures-labs/
#ArtificialIntelligence #DeepLearning #GenerativeAdversarialNetworks #MachineLearning #NeuralNetworks

Keras / TPU integration in Tensorflow 2.1 (unreleased)

Google Cloud Platform : https://github.com/GoogleCloudPlatform/training-data-analyst/blob/master/courses/fast-and-lean-data-science/README-TF2.1.md

#Keras #TPU #Tensorflow

Game Theory is a branch of mathematics used to model the strategic interaction between different players in a context with predefined rules and outcomes.
Game Theory can be applied in different ambit of Artificial Intelligence:
Multi-agent AI systems.
Imitation and Reinforcement Learning.
Adversary training in Generative Adversarial Networks (GANs).
Game Theory can also be used to describe many situations in our daily life and Machine Learning models
Game Theory can be divided into 5 main types of games:
Cooperative vs Non-Cooperative Games: In cooperative games, participants can establish alliances in order to maximise their chances to win the game (eg. negotiations). In non-cooperative games, participants can’t instead form alliances (eg. wars).
Symmetric vs Asymmetric Games: In a symmetric game all the participants have the same goals and just their strategies implemented in order to achieve them will determine who wins the game (eg. chess). In asymmetric games instead, the participants have different or conflicting goals.
Perfect vs Imperfect Information Games: In Perfect Information games all the players can see the other players moves (eg. chess). Instead, in Imperfect Information games, the other players' moves are hidden (eg. card games).
Simultaneous vs Sequential Games: In Simultaneous games, the different players can take actions concurrently. Instead in Sequential games, each player is aware of the other players' previous actions (eg. board games).
Zero-Sum vs Non-Zero Sum Games: In Zero Sum games, if a player gains something that causes a loss to the other players. In Non-Zero Sum games, instead, multiple players can take benefit of the gains of another player.
Different aspects of Game Theory are commonly used in Artificial Intelligence, I will now introduce you to the Nash Equilibrium, Inverse Game Theory, designing AI Agents environments, and give you some practical examples.
Generative Adversarial Networks (GANs)
Multi-Agents Reinforcement Learning (MARL)

https://towardsdatascience.com/game-theory-in-artificial-intelligence-57a7937e1b88

Grammarly AI: The sweet spot of deep learning and natural language processing

https://bdtechtalks.com/2019/10/17/grammarly-ai-assistant-grammar-checker/

16. Appendix: Mathematics for Deep Learning¶
https://d2l.ai/chapter_appendix_math/index.html

Geometry of Deep Learning

https://www.youtube.com/watch?v=mW96gzCPT_Q

Q8BERT: Quantized 8Bit BERT
Zafrir et al.: https://arxiv.org/abs/1910.06188
#NaturalLanguageProcessing #NLP #Transformer

How to Program UMAP from Scratch
And how to improve UMAP
By Nikolay Oskolkov : https://towardsdatascience.com/how-to-program-umap-from-scratch-e6eff67f55fe
#MachineLearning #DataScience #Bioinformatics