Many AI experts believe humanlike artificial general intelligence (AGI) is but a far-fetched dream, while others find their inspiration in the quest for AGI. Speaking at last November’s AI Frontiers Conference, OpenAI Founder and Research Director Ilya Sutskever said “We (OpenAI) have reviewed progress in the field over the past few years. Our conclusion is near-term AGI should be taken as a serious possibility.”
Today, respected scientific journal Nature boosted the case for AGI with a cover story on a new research paper, Towards artificial general intelligence with hybrid Tianjic chip architecture, which aims to stimulate AGI development by adopting generalized hardware platforms.
Typically, researchers have taken one of two paths in pursuit of AGI — proceeding either via computer science or via neuroscience. Each approach however requires its own unique and incompatible platforms, and this has stalled overarching AGI research and development. With an eye on closing that gap, researchers from Tsinghua University, Beijing Lynxi Technology, Beijing Normal University, Singapore Polytechnic University and University of California Santa Barbara have introduced the Tianjic chip. The revolutionary chip can adopt various core architectures, reconfigurable building blocks and so on, to accommodate both computer-science-based machine-learning algorithms and neuroscience-oriented schemes such as brain-inspired circuits.
A key innovation from the research team is Tianjic’s unified function core (FCore) which combines essential building blocks for both artificial neural networks and biologically networks — axon, synapse, dendrite and soma blocks. The 28-nm chip consists of 156 FCores, containing approximately 40,000 neurons and 10 million synapses in an area of 3.8×3.8 mm2.
Tianjic delivers an internal memory bandwidth of more than 610 gigabytes (GB) per second, and a peak performance of 1.28 tera operations per second (TOPS) per watt for running artificial neural networks. In the biologically-inspired spiking neural network mode, Tianjic achieves a peak performance of about 650 giga synaptic operations per second (GSOPS) per watt. The research team also showcased the superior performance of Tianjic compared to GPU, where the new chip achieves 1.6 – 100 times better throughput and 12 – 10000 times better power efficiency.
The research team designed a self-driving bicycle experiment to evaluate the chip’s capability for integrating multimodal information and making prompt decisions. Equipped with the Tianjic chip and IMU sensor, a camera, steering motor, driving motor, speed motor and battery, the bicycle was tasked with performing functions such as real-time object detection, tracking, voice-command recognition, riding over a speed bump, obstacle avoidance, balance control and decision making.
The research team developed a variety of neural networks (CNN, CANN, SNN and MLP networks) to enable each task. The models were pretrained and programmed onto the Tianjic chip, which can process the models in parallel and enable seamless on-chip communication across different models.
In experiments, the Tianjic-powered bicycle smoothly performed all assigned tasks, signaling a huge leap towards the acceleration of AGI development.

The research team also noted that “high spatiotemporal complexity can be generated by randomly introducing new variables into the environment in real time, such as different road conditions, noises, weather factors, multiple languages, more people and so on. By exploring solutions that allow adaptation to these environmental changes, issues critical to AGI — such as generalization, robustness and autonomous learning — can be examined.”

The research team told Chinese media they expect the Tianjic chip to be deployed in autonomous vehicles and smart robots. They have already started research on the next-generation chips and expect to close the R&D stage early next year.

Further information can be found in the paper Towards artificial general intelligence with hybrid Tianjic chip architecture.
https://syncedreview.com/2019/07/31/nature-cover-story-chinese-teams-tianjic-chip-bridges-machine-learning-and-neuroscience-in-pursuit-of-agi/

Modeling question asking using neural program generation
Ziyun Wang and Brenden M. Lake : https://arxiv.org/abs/1907.09899
#artificialintelligence #naturallanguageprocessing #reinforcementlearning

Landmark Detection in Low Resolution Faces with Semi-Supervised Learning. arxiv.org/abs/1907.13255

Wasserstein Robust Reinforcement Learning. arxiv.org/abs/1907.13196

AI-GAs: AI-generating algorithms, an alternate paradigm for producing general artificial intelligence

"We should keep in mind the grandeur of the task we are discussing, which is nothing short than the creation of an artificial intelligence smarter than humans. If we succeed, we arguably have also created life itself..."

By Jeff Clune : https://arxiv.org/abs/1905.10985

#ArtificialIntelligence #ArtificialGeneralIntelligence #MetaLearning

DeepMind: Using AI to give doctors a 48-hour head start on life-threatening illness

Blog: https://deepmind.com/blog/predicting-patient-deterioration/
Paper: https://www.nature.com/articles/s41586-019-1390-1

Three great lecture series on the brain


I know three great lecture series on neuroscience. Free for all available on YouTube. These series have let me study complex brain science without a background in STEM.

One on biological psychology and the brain

https://www.youtube.com/playlist?list=PLtXCbh6IFA7QCsei-t8WesusKi8I2LXUJ

And another on brain anatomy using real brain dissections

https://www.youtube.com/playlist?list=PLp9HSlEm97VXyQ32Uwjfz3dpmQ8nl63zJ

The last one focuses on cellular mechanisms of brain function.

https://www.youtube.com/channel/UCmB_Ytx1-QmOE52Ch5QjFIA

It's an amazing time that we live in time period where all these resources are free for all to use. If anybody knows anymore please post in comment section.

Facebook, Carnegie Mellon build first AI that beats pros in 6-player poker
Written by Noam Brown : https://ai.facebook.com/blog/pluribus-first-ai-to-beat-pros-in-6-player-poker
#ArtificialIntelligence #MachineLearning #Poker

The call for paper for the 2nd edition of the Latinx in AI workshop at
NeurIPS 2019 Conference is out!

Deadline is September 13th, 2019.

Submission instructions here https://www.latinxinai.org/neurips-2019

DROGON: A Causal Reasoning Framework for Future Trajectory Forecast

Chiho Choi, Abhishek Patil, Srikanth Malla : https://arxiv.org/abs/1908.00024

#ArtificialIntelligence #CausalReasoning #Robotics

Real-Time Voice Cloning

GitHub, by Corentin Jemine : https://github.com/CorentinJ/Real-Time-Voice-Cloning

#deeplearning #pytorch #tensorflow #naturallanguageprocessing

PostDoc for Machine Learning in Clinical Neurology



We are looking for a highly motivated and skilled postdoc to apply and develop machine learning methods in clinical neurology. In a collaboration with the Hertie Institute for Clinical Brain Research, the candidate will have several opportunities to work on the prediction of factors and progression of neurodegenerative diseases and epilepsy, using rich longitudinal clinical and molecular data.

The ideal candidate has a PhD in machine learning, physics, math, electrical engineering, or related fields, and a strong background in mathematics, machine learning, statistics, and ideally causal learning, with prior experience in working on life science/clinical data.

We offer a thriving and interactive environment in one of Europe’s leading location for machine learning, and the opportunity to directly work with clinicians and data from the large neurology section of Tübingen’s university hospital.

We particularly encourage women and other underrepresented groups in STEM fields to apply. Please check the How to apply section below.

PhD student in Machine Learning on Inductive Bias Transfer
We are looking for a PhD student to work on quantifying and transferring inductive biases between networks, or neuroscientific data and artificial neuronal networks.

The ideal candidate should have a masters degree in computer science, physics, math, electrical engineering, or related fields, and with a strong background in mathematics and programming. Previous experience in machine learning not necessary but advantageous.

Tübingen is one of the leading locations for machine learning in Europe, offering a scientific inspiring and open enviroment to develop the intelligent systems of tomorrow. The International Max Planck Graduate School for Intelligent Systems is Germany’s largest graduate program for machine learning and related topics.

We particularly encourage women and other underrepresented groups in STEM fields to apply. Please check the How to apply section below.

How to apply
We are always excited to receive applications from people to join our team. If you decide to reach out, please address the following points in your initial email (note that these lists are inclusive; PostDoc applicants have to submit all points below)

General
What are you applying for (lab rotation, research assistant, PhD position, postdoc position, etc.)?
What is your envisioned time frame (start date, end date)?
Do you bring own (partial) funding or will you need to be funded from here?
PhD position and beyond
Curriculum Vitae (clearly stating your education and scientific background)
Statement of motivation why you would like to join the lab (pdf, max 1/2 page)
Links to code you have written (e.g. github or bitbucket)
If applicable, at most two publications or pre-print you (co)-authored
PostDoc
A concise (pdf, max 1/2 page) “research proposal” on a project you would be interested working on. This is not binding but helps to get the conversation started. Applicants answering calls for a specfic project can ignore this.

https://sinzlab.org/openpositions.html

I want to pursue machine learning as a career but not sure if I am qualified. How can I test myself?


answer Andrew Ng :

You are qualified for a career in machine learning! Whatever your current level of knowledge, so long as you keep working hard and keep learning, you can become expert in machine learning and have a great career there.

To anyone interested in the field, start with learning to program. When you’ve mastered the basics of programming, consider the Machine Learning course (Machine Learning | Coursera), then the Deep Learning specialization (deeplearning.ai).

To go even further, read research papers (follow ML leaders on twitter to see what papers they’re excited about). Even better, try to replicate the results in the research papers. Trying to replicating others’ results is a one of the most effective but under-appreciated ways to get good at AI. Also consider activities like online ML competitions, academic conferences, and keep reading books/blogs/papers.

It’s really not a matter of whether you’re qualified do work in machine learning—I’m sure you are! It’s just a matter of getting the learnings to make yourself more and more qualified.

new paper from Andrew Ng

👇👇👇👇

Postdoc Artificial intelligence for ultrasound imaging at Eindhoven University of Technology, Netherlands

Offer Requirements
SPECIFIC REQUIREMENTS
We are looking for candidates who have:
A PhD degree in a relevant area.
Proven experience both in deep learning and in medical imaging.
The ability to contribute to cross-disciplinary collaborations.
Excellent written and oral communication skills in English.
A result-driven and proactive attitude.
The position is open and can be filled at any time, but no later than 1 January 2020.

https://www.marktechpost.com/job/postdoc-artificial-intelligence-for-ultrasound-imaging-at-eindhoven-university-of-technology-netherlands/