Identifying and eliminating bugs in learned predictive models
https://deepmind.com/blog/article/robust-and-verified-ai

Deep Learning model that good at Calculus. Nice.


https://openreview.net/forum?id=S1eZYeHFDS

AI solution about RealEstate in Singapore
They built on open-source geospatial features and were able to predict Singapore real estate prices with 87% accuracy (i.e., within an error margin of S$100).
They used a) OpenStreetMap (https://download.geofabrik.de/asia/malaysia-singapore-brunei.html)
b) Geomancer for geospatial features (https://stories.thinkingmachin.es/geomancer/)
Link: https://download.geofabrik.de/asia/malaysia-singapore-brunei.html

Extreme Language Model Compression with Optimal Subwords and Shared Projections
Zhao et al.: https://arxiv.org/abs/1909.11687
#neuralnetwork #bert #nlp

Mathematical Reasoning in Latent Space
Lee et al.: https://arxiv.org/pdf/1909.11851v1.pdf
#Mathematics #Reasoning #LatentSpace

Learning Pixel Representations for Generic Segmentation. https://arxiv.org/abs/1909.11735

Revisit Knowledge Distillation: a Teacher-free Framework. https://arxiv.org/abs/1909.11723

Augmenting the Pathology Lab: An Intelligent Whole Slide Image Classification System for... https://arxiv.org/abs/1909.11212

Scale-Equivariant Neural Networks with Decomposed Convolutional Filters. https://arxiv.org/abs/1909.11193

Neural reparameterization improves structural optimization
Hoyer et al.: https://arxiv.org/abs/1909.04240
#MachineLearning #NeuralNetworks #Optimization

Infinite-resolution image generation in the browser
Made by Erlend Gjesteland Ekern using CPPNs, GANs and TensorFlow.js : https://thispicturedoesnotexist.com
#CPPN #GAN #TensorflowJS

Representation Wars: Enacting an Armistice through Active Inference
https://www.researchgate.net/publication/333969028_Representation_Wars_Enacting_an_Armistice_through_Active_Inference

Artificial Design: Modeling Artificial Super Intelligence with Extended General Relativity and Universal Darwinism via Geometrization for Universal Design Automation
https://openreview.net/forum?id=SyxQ_TEFwS

GAMIN: An Adversarial Approach to Black-Box Model Inversion. https://arxiv.org/abs/1909.11835

#weekend_read
Check out this mind-blowing survey on HRL with some really-strong proven hypothesis. Maybe the best till date!
Paper-Title: WHY DOES HIERARCHY (SOMETIMES) WORK SO WELL IN REINFORCEMENT LEARNING?
Link to the paper: https://arxiv.org/pdf/1909.10618.pdf
#GoogleAI #UCB
Four Hypothesis: The four hypotheses may also be categorized as hierarchical training (H1 and H3) and hierarchical exploration (H2 and H4).
(H1) Temporally extended training. High-level actions correspond to multiple environment steps. To the high-level agent, episodes are effectively shorter. Thus, rewards are propagated faster and learning should improve.
(H2) Temporally extended exploration. Since high-level actions correspond to multiple environment steps, exploration in the high-level is mapped to environment exploration which is temporally correlated across steps. This way, an HRL agent explores the environment more efficiently. As a motivating example, the distribution associated with a random (Gaussian) walk is wider when the random noise is temporally correlated.
(H3) Semantic training. High-level actor and critic networks are trained with respect to semantically meaningful actions. These semantic actions are more correlated with future values, and thus easier to learn, compared to training with respect to the atomic actions of the environment. For example, in a robot navigation task it is easier to learn future values with respect to deltas in x-y coordinates rather than robot joint torques.
(H4) Semantic exploration. Exploration strategies (in the simplest case, random action noise) are applied to semantically meaningful actions and are thus more meaningful than the same strategies would be if applied to the atomic actions of the environment. For example, in a robot navigation task, it intuitively makes more sense to explore at the level of x-y coordinates rather than robot joint torques.
TL;DR: A large number of conclusions can be drawn based on empirical analysis. Here are few:-
In terms of the benefits of training, it is clear that training with respect to semantically meaningful abstract actions (H3) has a negligible effect on the success of HRL.
Moreover, temporally extended training (H1) is only important insofar as it enables the use of multi-step rewards, as opposed to training with respect to temporally extended actions.
The main and arguably most surprising, the benefit of the hierarchy is due to exploration. This is evidenced by the fact that temporally extended goal-reaching and agent-switching can enable non-hierarchical agents to solve tasks that otherwise can only be solved.
These results suggest that the empirical effectiveness of hierarchical agents simply reflects the improved exploration that these agents can attain.

Yoshua Bengio brainstorm with students. Very interesting discussion

[https://www.youtube.com/watch?v=g9V-MHxSCcs](https://www.youtube.com/watch?v=g9V-MHxSCcs)

‘Backronym’ May Help To Generate Ideas in Machine Learning by Visualizing Hundreds of Research Papers Together

Website: https://backronym.xyz/graph.html
Paper: https://arxiv.org/pdf/1908.01874v2.pdf
https://www.marktechpost.com/2019/09/29/backronym-may-help-to-generate-ideas-in-machine-learning-by-visualizing-hundreds-of-research-papers-together/

Our preliminary results for keyphrase generation using GANs. This is the first time GANs are used for generating keyphrases. [https://arxiv.org/pdf/1909.12229.pdf](https://t.co/BON7QWOsra?amp=1)

They work pretty well better than some of the popular models for generating keyphrases.

PostDoc position at Yale University

The Yale Center for Medical Informatics seeks a highly motivated Postdoctoral Associate in data science. Individuals with expertise in machine learning, deep learning, or natural language processing in the context of biomedical informatics are highly encouraged to apply. The position is focused on working with healthcare data including text data, administrative data, and image data to develop and test new approaches to study different healthcare pressing issues such as headaches phenotyping.
@ArtificialIntelligenceArticles

The successful candidate will work in close collaboration with the research team developers, participating in all phases of the research project: conceptualization, development of machine learning approaches; design and implementation; management and analysis; writing up results; and presenting findings. This two-year, full-time position (with the potential for extension) offers the opportunity to work alongside a prolific group of biomedical informatics researchers but requires creativity, excellent writing skills, and a strong interest in intellectual collaboration with a multi-disciplinary team. Position beginning Novemebr 2019, materials should be submitted no later than September 30, 2019.

The successful fellow will receive health insurance coverage and full access to the Yale University libraries.



Application Instructions

Candidates are requested to submit

· Curriculum Vitae or Resume

· 2 names of references

To: Recruiter ([email protected])
@ArtificialIntelligenceArticles
Yale University is an Affirmative Action/Equal Opportunity Employer and welcomes applications from women, members of minority groups, persons with disabilities and protected veterans.

Research Software Engineer in Machine Learning Applied to Finance, University of Oxford

Research Software Engineer in Machine Learning Applied to Finance
Department of Engineering Science, Oxford

We are seeking a full-time Research Software Engineer to work on applied Machine Learning and Software Development for Finance. The post involves the continued development of machine learning and financial simulation algorithms as well as liaison with industry partners (our collaborators in Man Group) to enable proof of concept deployment of algorithms on extensive data sets.

The postholder will have experience in producing software for data-centric machine learning. The role will focus as much on creating robust, reliable solutions as it will on innovation, suiting a candidate who is motivated by creating end to end solutions to address real-world problems.

You should possess a first degree in engineering, physics, computer science, mathematics, statistics or similar. Experience in practical applications of machine learning as well as expertise and experience in computer programming are essential. You should have a track record of experience and the ability to work well both independently and as part of a team.

You will be required to upload a covering letter/supporting statement, including a brief statement of research interests (describing how past experience and future plans fit with the advertised position), CV and the details of two referees as part of your online application. Only applications received before 12.00 midday on 24 October 2019 can be considered.


More information can be found


https://my.corehr.com/pls/uoxrecruit/erq_jobspec_version_4.display_form

TensorFlow 2.0 has been released! @ArtificialIntelligenceArticles
Release 2.0.0: https://github.com/tensorflow/tensorflow/releases/tag/v2.0.0
#ArtificialIntelligence #DeepLearning #TensorFlow