Category Archives: Machine Learning
The way we train AI is fundamentally flawed – MIT Technology Review
For example, they trained 50 versions of an image recognition model on ImageNet, a dataset of images of everyday objects. The only difference between training runs were the random values assigned to the neural network at the start. Yet despite all 50 models scoring more or less the same in the training testsuggesting that they were equally accuratetheir performance varied wildly in the stress test.
The stress test used ImageNet-C, a dataset of images from ImageNet that have been pixelated or had their brightness and contrast altered, and ObjectNet, a dataset of images of everyday objects in unusual poses, such as chairs on their backs, upside-down teapots, and T-shirts hanging from hooks. Some of the 50 models did well with pixelated images, some did well with the unusual poses; some did much better overall than others. But as far as the standard training process was concerned, they were all the same.
The researchers carried out similar experiments with two different NLP systems, and three medical AIs for predicting eye disease from retinal scans, cancer from skin lesions, and kidney failure from patient records. Every system had the same problem: models that should have been equally accurate performed differently when tested with real-world data, such as different retinal scans or skin types.
We might need to rethink how we evaluate neural networks, says Rohrer. It pokes some significant holes in the fundamental assumptions we've been making.
DAmour agrees. The biggest, immediate takeaway is that we need to be doing a lot more testing, he says. That wont be easy, however. The stress tests were tailored specifically to each task, using data taken from the real world or data that mimicked the real world. This is not always available.
Some stress tests are also at odds with each other: models that were good at recognizing pixelated images were often bad at recognizing images with high contrast, for example. It might not always be possible to train a single model that passes all stress tests.
One option is to design an additional stage to the training and testing process, in which many models are produced at once instead of just one. These competing models can then be tested again on specific real-world tasks to select the best one for the job.
Thats a lot of work. But for a company like Google, which builds and deploys big models, it could be worth it, says Yannic Kilcher, a machine-learning researcher at ETH Zurich. Google could offer 50 different versions of an NLP model and application developers could pick the one that worked best for them, he says.
DAmour and his colleagues dont yet have a fix but are exploring ways to improve the training process. We need to get better at specifying exactly what our requirements are for our models, he says. Because often what ends up happening is that we discover these requirements only after the model has failed out in the world.
Getting a fix is vital if AI is to have as much impact outside the lab as it is having inside. When AI underperforms in the real-world it makes people less willing to want to use it, says co-author Katherine Heller, who works at Google on AI for healthcare: We've lost a lot of trust when it comes to the killer applications, thats important trust that we want to regain.
Read the original post:
The way we train AI is fundamentally flawed - MIT Technology Review
DIY Camera Uses Machine Learning to Audibly Tell You What it Sees – PetaPixel
Adafruit Industries has created a machine learning camera built with the Raspberry Pi that can identify objects extremely quickly and audibly tell you what it sees. The group has listed all the necessary parts you need to build the device at home.
The camera is based on Adafruits BrainCraft HAT add-on for the Raspberry Pi 4, and uses TensorFlow Lite object recognition software to be able to recognize what it is seeing. According to Adafruits website, its compatible with both the 8-megapixel Pi camera and the 12.3-megapixel interchangeable lens version of module.
While interesting on its own, DIY Photography makes a solid point by explaining a more practical use case for photographers:
You could connect a DSLR or mirrorless camera from its trigger port into the Pis GPIO pins, or even use a USB connection with something like gPhoto, to have it shoot a photo or start recording video when it detects a specific thing enter the frame.
A camera that is capable of recognizing what it is looking at could be used to only take a photo when a specific object, animal, or even a person comes into the frame. That would mean it could have security system or wildlife monitoring applications. Whenever you might wish your camera knew what it was looking at, this kind of technology would make that a reality.
You can find all the parts you will need to build your own version of this device on Adafruits website here. They also have published an easy machine learning guide for the Raspberry Pi as well as a guide on running TensorFlow Lite.
(via DPReview and DIY Photography)
See the article here:
DIY Camera Uses Machine Learning to Audibly Tell You What it Sees - PetaPixel
Machine Learning Predicts How Cancer Patients Will Respond to Therapy – HealthITAnalytics.com
November 18, 2020 -A machine learning algorithm accurately determined how well skin cancer patients would respond to tumor-suppressing drugs in four out of five cases, according to research conducted by a team from NYU Grossman School of Medicine and Perlmutter Cancer Center.
The study focused on metastatic melanoma, a disease that kills nearly 6,800 Americans each year. Immune checkpoint inhibitors, which keep tumors from shutting down the immune systems attack on them, have been shown to be more effective than traditional chemotherapies for many patients with melanoma.
However, half of patients dont respond to these immunotherapies, and these drugs are expensive and often cause side effects in patients.
While immune checkpoint inhibitors have profoundly changed the treatment landscape in melanoma, many tumors do not respond to treatment, and many patients experience treatment-related toxicity, said corresponding study authorIman Osman, medical oncologist in the Departments of Dermatology and Medicine (Oncology) at New York University (NYU) Grossman School of Medicine and director of the Interdisciplinary Melanoma Program at NYU Langones Perlmutter Cancer Center.
An unmet need is the ability to accurately predict which tumors will respond to which therapy. This would enable personalized treatment strategies that maximize the potential for clinical benefit and minimize exposure to unnecessary toxicity.
READ MORE: How Social Determinants Data Can Enhance Machine Learning Tools
Researchers set out to develop a machine learning model that could help predict a melanoma patients response to immune checkpoint inhibitors. The team collected 302 images of tumor tissue samples from 121 men and women treated for metastatic melanoma with immune checkpoint inhibitors at NYU Langone hospitals.
They then divided these slides into 1.2 million portions of pixels, the small bits of data that make up images. These were fed into the machine learning algorithm along with other factors, such as the severity of the disease, which kind of immunotherapy regimen was used, and whether a patient responded to the treatment.
The results showed that the machine learning model achieved an AUC of 0.8 in both the training and validation cohorts, and was able to predict which patients with a specific type of skin cancer would respond well to immunotherapies in four out of five cases.
Our findings reveal that artificial intelligence is a quick and easy method of predicting how well a melanoma patient will respond to immunotherapy, said study first author Paul Johannet, MD, a postdoctoral fellow at NYU Langone Health and its Perlmutter Cancer Center.
Researchers repeated this process with 40 slides from 30 similar patients at Vanderbilt University to determine whether the results would be similar at a different hospital system that used different equipment and sampling techniques.
READ MORE: Simple Machine Learning Method Predicts Cirrhosis Mortality Risk
A key advantage of our artificial intelligence program over other approaches such as genetic or blood analysis is that it does not require any special equipment, said study co-author Aristotelis Tsirigos, PhD, director of applied bioinformatics laboratories and clinical informatics at the Molecular Pathology Lab at NYU Langone.
The team noted that aside from the computer needed to run the program, all materials and information used in the Perlmutter technique are a standard part of cancer management that most, if not all, clinics use.
Even the smallest cancer center could potentially send the data off to a lab with this program for swift analysis, said Osman.
The machine learning method used in the study is also more streamlined than current predictive tools, such as analyzing stool samples or genetic information, which promises to reduce treatment costs and speed up patient wait times.
Several recent attempts to predict immunotherapy responses do so with robust accuracy but use technologies, such as RNA sequencing, that are not readily generalizable to the clinical setting, said corresponding study authorAristotelis Tsirigos, PhD, professor in the Institute for Computational Medicine at NYU Grossman School of Medicine and member of NYU Langones Perlmutter Cancer Center.
READ MORE: Machine Learning Forecasts Prognosis of COVID-19 Patients
Our approach shows that responses can be predicted using standard-of-care clinical information such as pre-treatment histology images and other clinical variables.
However, researchers also noted that the algorithm is not yet ready for clinical use until they can boost the accuracy from 80 percent to 90 percent and test the algorithm at more institutions. The research team plans to collect more data to improve the performance of the model.
Even at its current level of accuracy, the model could be used as a screening method to determine which patients across populations would benefit from more in-depth tests before treatment.
There is potential for using computer algorithms to analyze histology images and predict treatment response, but more work needs to be done using larger training and testing datasets, along with additional validation parameters, in order to determine whether an algorithm can be developed that achieves clinical-grade performance and is broadly generalizable, said Tsirigos.
There is data to suggest that thousands of images might be needed to train models that achieve clinical-grade performance.
See the original post:
Machine Learning Predicts How Cancer Patients Will Respond to Therapy - HealthITAnalytics.com
This New Machine Learning Tool Might Stop Misinformation – Digital Information World
Misinformation has always been a problem, but the combination of widespread social media as well as a loose definition of what can be seen as factual truth in recent times has lead to a veritable explosion in misinformation over the course of the past few years. The problem is so dire that in a lot of cases websites are made specifically because of the fact that this is the sort of thing that could potentially end up allowing misinformation to spread more easily, and this is a problem that might just have been addressed by a new machine learning tool.
This machine learning tool was developed by researchers at UCL, Berkeley and Cornell will be able to detect domain registration data and use this to ascertain whether the URL is legitimate or if it has been made specifically to legitimize a certain piece of information that people might be trying to spread around. A couple of other factors also come into play here. For example, if the identity of the person that registered the domain is private, this might be a sign that the site is not legitimate. The timing of the domain registration matters to. If it was done around the time a major news event broke out, such as the recent US presidential election, this is also a negative sign.
With all of that having been said and out of the way, it is important to note that this new machine learning tool has a pretty impressive success rate of about 92%, which is the proportion of fake domains it was able to discover. Being able to tell whether or not a news source is legitimate or whether it is direct propaganda is useful because of the fact that it can help reduce the likelihood that people might just end up taking the misinformation seriously.
Original post:
This New Machine Learning Tool Might Stop Misinformation - Digital Information World
Fujitsu, AIST and RIKEN Achieve Unparalleled Speed on MLPerf HPC Machine Learning Processing Benchmark – HPCwire
TOKYO, Nov 19, 2020 Fujitsu, the National Institute of Advanced Industrial Science and Technology (AIST), and RIKEN today announced a performance milestone in supercomputing, achieving the highest performance and claiming the ranking positions on the MLPerf HPC benchmark. The MLPerf HPC benchmark measures large-scale machine learning processing on a level requiring supercomputers and the parties achieved these outcomes leveraging approximately half of the AI-Bridging Cloud Infrastructure (ABCI) supercomputer system, operated by AIST, and about 1/10 of the resources of the supercomputer Fugaku, which is currently under joint development by RIKEN and Fujitsu.
Utilizing about half the computing resources of its system, ABCI achieved processing speeds 20 times faster than other GPU-type systems. That is the highest performance among supercomputers based on GPUs, computing devices specialized in deep learning. Similarly, about 1/10 of Fugaku was utilized to set a record for CPU-type supercomputers consisting of general-purpose computing devices only, achieving a processing speed 14 times faster than that of other CPU-type systems.
The results were presented as MLPerf HPC v0.7 on November 18th (November 19th Japan Time) at the 2020 International Conference for High Performance Computing, Networking, Storage, and Analysis (SC20) event, which is currently being held online.
Background
MLPerf HPC is a performance competition in two benchmark programs: CosmoFlow(2), which predicts cosmological parameters, and DeepCAM, which identifies abnormal weather phenomena. The ABCI ranked first in metrics of all registered systems in the CosmoFlow benchmark program, with about half of the whole ABCI system, and Fugaku ranked second with measurement of about 1/10 of the whole system. The ABCI system delivered 20 times the performance of the other GPU types, while Fugaku delivered 14 times the performance of the other CPU types. ABCI achieved first place amongst all registered systems in the DeepCAM benchmark program as well, also with about half of the system. In this way, ABCI and Fugaku overwhelmingly dominated the top positions, demonstrating the superior technological capabilities of Japanese supercomputers in the field of machine learning.
Fujitsu, AIST, RIKEN and Fujitsu Laboratories Limited will release the software stacks including the library and the AI framework which accelerate the large-scale machine learning process developed for this measurement to the public. This move will make it easier to use large-scale machine learning with supercomputers, while its use in analyzing simulation results is anticipated to contribute to the detection of abnormal weather phenomena and to new discoveries in astrophysics. As a core platform for building Society 5.0, it will also contribute to solve social and scientific issues, as it is expected to expand to applications such as the creation of general-purpose language models that require enormous computational performance.
About MLPerf HPC
MLPerf is a machine learning benchmark community established in May 2018 for the purpose of creating a performance list of systems running machine learning applications. MLPerf developed MLPerf HPC as a new machine learning benchmark to evaluate the performance of machine learning calculations using supercomputers. It is used for supercomputers around the world and is expected to become a new industry standard. MLPerf HPC v0.7 evaluated performance on two real applications, CosmoFlow and DeepCAM, to measure large-scale machine learning performance requiring the use of a supercomputer.
All measurement data are available on the following website: https://mlperf.org/
Comments from the Partners
Fujitsu, Executive Director, Naoki Shinjo: The successful construction and optimization of the software stack for large-scale deep learning processing, executed in close collaboration with AIST, RIKEN, and many other stakeholders made this achievement a reality, helping us to successfully claim the top position in the MLPerf HPC benchmark in an important milestone for the HPC community. I would like to express my heartfelt gratitude to all concerned for their great cooperation and support. We are confident that these results will pave the way for the use of supercomputers for increasingly large-scale machine learning processing tasks and contribute to many research and development projects in the future, and we are proud that Japans research and development capabilities will help lead global efforts in this field.
Hirotaka Ogawa, Principal Research Manager, Artificial Intelligence Research Center, AIST: ABCI was launched on August 1, 2018 as an open, advanced, and high-performance computing infrastructure for the development of artificial intelligence technologies in Japan. Since then, it has been used in industry-academia-government collaboration and by a diverse range of businesses, to accelerate R&D and verification of AI technologies that utilize high computing power, and to advance social utilization of AI technologies. The overwhelming results of MLPerf HPC, the benchmark for large-scale machine learning processing, showed the world the high level of technological capabilities of Japans industry-academia-government collaboration. AISTs Artificial Intelligence Research Center is promoting the construction of large-scale machine learning models with high versatility and the development of its application technologies, with the aim of realizing easily-constructable AI. We expect that the results of this time will be utilized in such technological development.
Satoshi Matsuoka, Director General, RIKEN Center for Computational Science: In this memorable first MLPerf HPC, Fugaku, Japans top CPU supercomputer, along with AISTs ABCI, Japans top GPU supercomputer, exhibited extraordinary performance and results, serving as a testament to Japans ability to compete at an exceptional level on the global stage in the area of AI research and development. I only regret that we couldnt achieve the overwhelming performance as we did for HPL-AI to be compliant with inaugural regulations for MLPerf HPC benchmark. In the future, as we continue to further improve the performance on Fugaku, we will make ongoing efforts to take advantage of Fugakus super large-scale environment in the area of high-performance deep learning in cooperation with various stakeholders.
About Fujitsu
Fujitsu is a leading Japanese information and communication technology (ICT) company offering a full range of technology products, solutions and services. Approximately 130,000 Fujitsu people support customers in more than 100 countries. We use our experience and the power of ICT to shape the future of society with our customers. Fujitsu Limited (TSE:6702) reported consolidated revenues of 3.9 trillion yen (US$35 billion) for the fiscal year ended March 31, 2020. For more information, please see http://www.fujitsu.com.
About National Institute of Advanced Industrial Science & Technology (AIST)
AIST is the largest public research institute established in 1882 in Japan. The research fields of AIST covers all industrial sciences, e.g., electronics, material science, life science, metrology, etc. Our missions are bridging the gap between basic science and industrialization and solving social problems facing the world. we prepare several open innovation platforms to contribute to these missions, where researchers in companies, university professors, graduated students, as well as AIST researchers, get together to achieve our missions. The open innovation platform established recently is The Global Zero Emission Research Center which contributes to achieving a zero-emission society collaborating with foreign researches.https://www.aist.go.jp/index_en.html
About RIKEN Center for Computational Science
RIKEN is Japans largest comprehensive research institution renowned for high-quality research in a diverse range of scientific disciplines. Founded in 1917 as a private research foundation in Tokyo, RIKEN has grown rapidly in size and scope, today encompassing a network of world-class research centers and institutes across Japan including the RIKEN Center for Computational Science (R-CCS), the home of the supercomputer Fugaku. As the leadership center of high-performance computing, the R-CCS explores the Science of computing, by computing, and for computing. The outcomes of the exploration the technologies such as open source software are its core competence. The R-CCS strives to enhance the core competence and to promote the technologies throughout the world.
Source: Fujitsu
SVG Tech Insight: Increasing Value of Sports Content Machine Learning for Up-Conversion HD to UHD – Sports Video Group
This fall SVG will be presenting a series of White Papers covering the latest advancements and trends in sports-production technology. The full series of SVGs Tech Insight White Papers can be found in the SVG Fall SportsTech Journal HERE.
Following the height of the 2020 global pandemic, live sports are starting to re-emerge worldwide albeit predominantly behind closed doors. For the majority of sports fans, video is the only way they can watch and engage with their favorite teams or players. This means the quality of the viewing experience itself has become even more critical.
With UHD being adopted by both households and broadcasters around the world, there is a marked expectation around visual quality. To realize these expectations in the immediate term, it will be necessary for some years to up-convert from HD to UHD when creating 4K UHD sports channels and content.
This is not so different from the early days of HD, where SD sporting related content had to be up-converted to HD. In the intervening years, however, machine learning as a technology has progressed sufficiently to be a serious contender for performing better up-conversions than with more conventional techniques, specifically designed to work for TV content.
Ideally, we want to process HD content into UHD with a simple black box arrangement.
The problem with conventional up-conversion, though, is that it does not offer an improved resolution, so does not fully meet the expectations of the viewer at home watching on a UHD TV. The question, therefore, becomes: can we do better for the sports fan? If so, how?
UHD is a progressive scan format, with the native TV formats being 38402160, known as 2160p59.64 (usually abbreviated to 2160p60) or 2160p50. The corresponding HD formats, with the frame/field rates set by region, are either progressive 1280720 (720p60 or 720p50) or interlaced 19201080 (1080i30 or 1080i25).
Conversion from HD to UHD for progressive images at the same rate is fairly simple. It can be achieved using spatial processing only. Traditionally, this might typically use a bi-cubic interpolation filter, (a 2-dimensional interpolation commonly used for photographic image scaling.) This uses a grid of 44 source pixels and interpolates intermediate locations in the center of the grid. The conversion from 1280720 to 38402160 requires a 3x scaling factor in each dimension and is almost the ideal case for an upsampling filter.
These types of filters can only interpolate, resulting in an image that is a better result than nearest-neighbor or bi-linear interpolation, but does not have the appearance of being a higher resolution.
Machine Learning (ML) is a technique whereby a neural network learns patterns from a set of training data. Images are large, and it becomes unfeasible to create neural networks that process this data as a complete set. So, a different structure is used for image processing, known as Convolutional Neural Networks (CNNs). CNNs are structured to extract features from the images by successively processing subsets from the source image and then processes the features rather than the raw pixels.
Up-conversion process with neural network processing
The inbuilt non-linearity, in combination with feature-based processing, mean CNNs can invent data not in the original image. In the case of up-conversion, we are interested in the ability to create plausible new content that was not present in the original image, but that doesnt modify the nature of the image too much. The CNN used to create the UHD data from the HD source is known as the Generator CNN.
When input source data needs to be propagated through the whole chain, possibly with scaling involved, then a specific variant of a CNN known as a Residual Network (ResNet) is used. A ResNet has a number of stages, each of which includes a contribution from a bypass path that carries the input data. For this study, a ResNet with scaling stages towards the end of the chain was used as the Generator CNN.
For the Generator CNN to do its job, it must be trained with a set of known data patches of reference images and a comparison is made between the output and the original. For training, the originals are a set of high-resolution UHD images, down-sampled to produce HD source images, then up-converted and finally compared to the originals.
The difference between the original and synthesized UHD images is calculated by the compare function with the error signal fed back to the Generator CNN. Progressively, the Generator CNN learns to create an image with features more similar to original UHD images.
The training process is dependent on the data set used for training, and the neural network tries to fit the characteristics seen during training onto the current image. This is intriguingly illustrated in Googles AI Blog [1], where a neural network presented with a random noise pattern introduces shapes like the ones used during training. It is important that a diverse, representative content set is used for training. Patches from about 800 different images were used for training during the process of MediaKinds research.
The compare function affects the way the Generator CNN learns to process the HD source data. It is easy to calculate a sum of absolute differences between original and synthesized. This causes an issue due to training set imbalance; in this case, the imbalance is that real pictures have large proportions with relatively little fine detail, so the data set is biased towards regenerating a result like that which is very similar to the use of a bicubic interpolation filter.
This doesnt really achieve the objective of creating plausible fine detail.
Generative Adversarial Neural Networks (GANs) are a relatively new concept [2], where a second neural network, known as the Discriminator CNN, is used and is itself trained during the training process of the Generator CNN. The Discriminator CNN learns to detect the difference between features that are characteristic of original UHD images and synthesized UHD images. During training, the Discriminator CNN sees either an original UHD image or a synthesized UHD image, with the detection correctness fed back to the discriminator and, if the image was a synthesized one, also fed back to the Generator CNN.
Each CNN is attempting to beat the other: the Generator by creating images that have characteristics more like originals, while the Discriminator becomes better at detecting synthesized images.
The result is the synthesis of feature details that are characteristic of original UHD images.
With a GAN approach, there is no real constraint to the ability of the Generator CNN to create new detail everywhere. This means the Generator CNN can create images that diverge from the original image in more general ways. A combination of both compare functions can offer a better balance, retaining the detail regeneration, but also limiting divergence. This produces results that are subjectively better than conventional up-conversion.
Conversion from 1080i60 to 2160p60 is necessarily more complex than from 720p60. Starting from 1080i, there are three basic approaches to up-conversion:
Training data is required here, which must come from 2160p video sequences. This enables a set of fields to be created, which are then downsampled, with each field coming from one frame in the original 2160p sequence, so the fields are not temporally co-located.
Surprisingly, results from field-based up-conversion tended to be better than using de-interlaced frame conversion, despite using sophisticated motion-compensated de-interlacing: the frame-based conversion being dominated by the artifacts from the de-interlacing process. However, it is clear that potentially useful data from the opposite fields did not contribute to the result, and the field-based approach missed data that could produce a better result.
A solution to this is to use multiple fields data as the source data directly into a modified Generator CNN, letting the GAN learn how best to perform the deinterlacing function. This approach was adopted and re-trained with a new set of video-based data, where adjacent fields were also provided.
This led to both high visual spatial resolution and good temporal stability. These are, of course, best viewed as a video sequence, however an example of one frame from a test sequence shows the comparison:
Comparison of a sample frame from different up-conversion techniques against original UHD
Up-conversion using a hybrid GAN with multiple fields was effective across a range of content, but is especially relevant for the visual sports experience to the consumer. This offers a realistic means by which content that has more of the appearance of UHD can be created from both progressive and interlaced HD source, which in turn can enable an improved experience for the fan at home when watching a sports UHD channel.
1 A. Mordvintsev, C. Olah and M. Tyka, Inceptionism: Going Deeper into Neural Networks, 2015. [Online]. Available: https://ai.googleblog.com/2015/06/inceptionism-going-deeper-into-neural.html
2 I. e. a. Goodfellow, Generative Adversarial Nets, Neural Information Processing Systems Proceedings, vol. 27, 2014.
Continued here:
SVG Tech Insight: Increasing Value of Sports Content Machine Learning for Up-Conversion HD to UHD - Sports Video Group
SiMa.ai Adopts Arm Technology to Deliver a Purpose-built Heterogeneous Machine Learning Compute Platform for the Embedded Edge – Design and Reuse
Licensing agreement enables machine learning intelligence with best-in-class performance and power for robotics, surveillance, autonomous, and automotive applications
SAN JOSE, Calif.-- November 18, 2020 -- SiMa.ai, the machine learning company enabling high performance compute at the lowest power, today announced the adoption of low-power Arm compute technology to build its purpose-built Machine Learning SoC (MLSoC) platform. The licensing of this technology brings machine learning intelligence with best-in-class performance and power to a broad set of embedded edge applications including robotics, surveillance, autonomous, and automotive.
SiMa.ai is adopting Arm Cortex-A and Cortex-M processors optimized for power, throughput efficiency, and safety-critical tasks. In addition, SiMa.ai is leveraging a combination of widely used open-source machine learning frameworks from Arms vast ecosystem, to allow software to seamlessly enable machine learning for legacy applications at the embedded edge.
Arm is the industry leader in energy-efficient processor design and advanced computing, said Krishna Rangasayee, founder and CEO of SiMa.ai. The integration of SiMa.ai's high performance and low power machine learning accelerator with Arm technology accelerates our progress in bringing our MLSoC to the market, creating new solutions underpinned by industry-leading IP, the broad Arm ecosystem, and world-class support from its field and development teams."
From autonomous systems to smart cities, the applications enabled by ML at the edge are delivering increased functionality, leading to more complex device requirements, said Dipti Vachani, senior vice president and general manager, Automotive and IoT Line of Business at Arm. SiMa.ai is innovating on top of Arms foundational IP to create a unique low power ML SoC that will provide intelligence to the next generation of embedded edge use cases.
SiMa.ai is strategically leveraging Arm technology to deliver its unique Machine Learning SoC. This includes:
About SiMa.ai
SiMa.ai is a machine learning company enabling high performance compute at the lowest power. Initially focused on solutions for computer vision applications at the embedded edge, the company is led by a team of technology experts committed to delivering the industrys highest frames per second per watt solution to its customers. To learn more, visit http://www.sima.ai.
See the rest here:
SiMa.ai Adopts Arm Technology to Deliver a Purpose-built Heterogeneous Machine Learning Compute Platform for the Embedded Edge - Design and Reuse
Machine learning removes bias from algorithms and the hiring process – PRNewswire
Arena Analytics' Chief Data Scientist unveils a cutting edge technique that removes latent bias from algorithmic models.
Currently, the primary methods of reducing the impact of bias on models has been limited to adjusting input data or adjust models after-the-fact to ensure there is no disparate impact.
Recent reporting from the Wall Street Journal confirmed these as the most recent advances, concluding, "It's really up to the software engineers and leaders of the company to figure out how to fix it [or] go into the algorithm and tweak some of the main factors it considers in making its decisions."
For several years, Arena Analytics was also limited to these approaches, but that all changed 9 months ago. Up until then, Arena removed all data from the models that could correlate to protected classifications and then measured demographic parity.
"These efforts brought us in line with EEOC compliance thresholds - also known as the or 80% rule," explains Myra Norton, President/COO of Arena. "But we've always wanted to go further than a compliance threshold.We've wanted to surface a MORE diverse slate of candidates for every role in a client organization.And that's exactly what we've accomplished, now surpassing 95% in our representation of different classifications."
Chief Data Scientist Patrick Hagerty will explain at MLConf the way he and his team have leveraged techniques known asadversarial networks,an aspect of Generative Adversarial Networks (GAN's), tools that pit one algorithm against another.
"Arena's primary model predicts the outcomes our clients want, and Model Two is a Discriminator designed to predict a classification," says Hagerty. "The Discriminator attempts to detect the race, gender, background, and any other protected class data of a person. This causes the Predictor to adjust and optimize while eliminating correlations with the classifications the Discriminator is detecting."
Arena trained models to do this until achieving what's known as the Nash Equilibrium. This is the point at which the predictor and discriminator have reached peak optimization.
Arena's technology has helped industrious individuals find a variety of jobs - from RNs to medtechs, caregivers to cooks, concierge to security. Job candidates who Arena predicted for success include veterans with no prior experience in healthcare or senior/assisted living, recent high school graduates whose plans to work while attending college were up-ended, and former hospitality sector employees who decided to apply their dining service expertise to a new setting.
"We succeeded in our intent to reduce bias and diversify the workforce, but what surprised us was the impact this approach had on our core predictions. Data once considered unusable, such as commuting distance, we can now analyze because we've removed the potentially-associated protected-class-signal," says Michael Rosenbaum, Arena's founder and CEO. "As a result, our predictions are stronger AND we surface a more diverse slate of candidates across multiple spectrums. Our clients can now use their talent acquisition function to really support and lead out front on Diversity and Inclusion."
About Arena (https://www.arena.io/) applies predictive analytics and machine learning to solve talent acquisition challenges. Learning algorithms analyze a large amount of data topredict with high levels of accuracy the likelihood of different outcomes occurring, such as someone leaving, being engaged, having excellent attendance, and more. By revealing each individual's likely outcomes in specific positions, departments, and locations, Arena is transforming the labor market from one based on perception and unconscious bias, to one based on outcomes. Arena is currently growing dramatically within the healthcare and hospitality industry and expanding its offerings to other people intensive industries. For more information contact [emailprotected]arena.io
SOURCE Arena
See the original post here:
Machine learning removes bias from algorithms and the hiring process - PRNewswire
Using machine learning to track the pandemic’s impact on mental health – MIT News
Dealing with a global pandemic has taken a toll on the mental health of millions of people. A team of MIT and Harvard University researchers has shown that they can measure those effects by analyzing the language that people use to express their anxiety online.
Using machine learning to analyze the text of more than 800,000 Reddit posts, the researchers were able to identify changes in the tone and content of language that people used as the first wave of the Covid-19 pandemic progressed, from January to April of 2020. Their analysis revealed several key changes in conversations about mental health, including an overall increase in discussion about anxiety and suicide.
We found that there were these natural clusters that emerged related to suicidality and loneliness, and the amount of posts in these clusters more than doubled during the pandemic as compared to the same months of the preceding year, which is a grave concern, says Daniel Low, a graduate student in the Program in Speech and Hearing Bioscience and Technology at Harvard and MIT and the lead author of the study.
The analysis also revealed varying impacts on people who already suffer from different types of mental illness. The findings could help psychiatrists, or potentially moderators of the Reddit forums that were studied, to better identify and help people whose mental health is suffering, the researchers say.
When the mental health needs of so many in our society are inadequately met, even at baseline, we wanted to bring attention to the ways that many people are suffering during this time, in order to amplify and inform the allocation of resources to support them, says Laurie Rumker, a graduate student in the Bioinformatics and Integrative Genomics PhD Program at Harvard and one of the authors of the study.
Satrajit Ghosh, a principal research scientist at MITs McGovern Institute for Brain Research, is the senior author of the study, which appears in the Journal of Internet Medical Research. Other authors of the paper include Tanya Talkar, a graduate student in the Program in Speech and Hearing Bioscience and Technology at Harvard and MIT; John Torous, director of the digital psychiatry division at Beth Israel Deaconess Medical Center; and Guillermo Cecchi, a principal research staff member at the IBM Thomas J. Watson Research Center.
A wave of anxiety
The new study grew out of the MIT class 6.897/HST.956 (Machine Learning for Healthcare), in MITs Department of Electrical Engineering and Computer Science. Low, Rumker, and Talkar, who were all taking the course last spring, had done some previous research on using machine learning to detect mental health disorders based on how people speak and what they say. After the Covid-19 pandemic began, they decided to focus their class project on analyzing Reddit forums devoted to different types of mental illness.
When Covid hit, we were all curious whether it was affecting certain communities more than others, Low says. Reddit gives us the opportunity to look at all these subreddits that are specialized support groups. Its a really unique opportunity to see how these different communities were affected differently as the wave was happening, in real-time.
The researchers analyzed posts from 15 subreddit groups devoted to a variety of mental illnesses, including schizophrenia, depression, and bipolar disorder. They also included a handful of groups devoted to topics not specifically related to mental health, such as personal finance, fitness, and parenting.
Using several types of natural language processing algorithms, the researchers measured the frequency of words associated with topics such as anxiety, death, isolation, and substance abuse, and grouped posts together based on similarities in the language used. These approaches allowed the researchers to identify similarities between each groups posts after the onset of the pandemic, as well as distinctive differences between groups.
The researchers found that while people in most of the support groups began posting about Covid-19 in March, the group devoted to health anxiety started much earlier, in January. However, as the pandemic progressed, the other mental health groups began to closely resemble the health anxiety group, in terms of the language that was most often used. At the same time, the group devoted to personal finance showed the most negative semantic change from January to April 2020, and significantly increased the use of words related to economic stress and negative sentiment.
They also discovered that the mental health groups affected the most negatively early in the pandemic were those related to ADHD and eating disorders. The researchers hypothesize that without their usual social support systems in place, due to lockdowns, people suffering from those disorders found it much more difficult to manage their conditions. In those groups, the researchers found posts about hyperfocusing on the news and relapsing back into anorexia-type behaviors since meals were not being monitored by others due to quarantine.
Using another algorithm, the researchers grouped posts into clusters such as loneliness or substance use, and then tracked how those groups changed as the pandemic progressed. Posts related to suicide more than doubled from pre-pandemic levels, and the groups that became significantly associated with the suicidality cluster during the pandemic were the support groups for borderline personality disorder and post-traumatic stress disorder.
The researchers also found the introduction of new topics specifically seeking mental health help or social interaction. The topics within these subreddit support groups were shifting a bit, as people were trying to adapt to a new life and focus on how they can go about getting more help if needed, Talkar says.
While the authors emphasize that they cannot implicate the pandemic as the sole cause of the observed linguistic changes, they note that there was much more significant change during the period from January to April in 2020 than in the same months in 2019 and 2018, indicating the changes cannot be explained by normal annual trends.
Mental health resources
This type of analysis could help mental health care providers identify segments of the population that are most vulnerable to declines in mental health caused by not only the Covid-19 pandemic but other mental health stressors such as controversial elections or natural disasters, the researchers say.
Additionally, if applied to Reddit or other social media posts in real-time, this analysis could be used to offer users additional resources, such as guidance to a different support group, information on how to find mental health treatment, or the number for a suicide hotline.
Reddit is a very valuable source of support for a lot of people who are suffering from mental health challenges, many of whom may not have formal access to other kinds of mental health support, so there are implications of this work for ways that support within Reddit could be provided, Rumker says.
The researchers now plan to apply this approach to study whether posts on Reddit and other social media sites can be used to detect mental health disorders. One current project involves screening posts in a social media site for veterans for suicide risk and post-traumatic stress disorder.
The research was funded by the National Institutes of Health and the McGovern Institute.
Continued here:
Using machine learning to track the pandemic's impact on mental health - MIT News
AI Recognizes COVID-19 in the Sound of a Cough Machine Learning Times – The Predictive Analytics Times
Originally published in IEEE Spectrum, Nov 4, 2020.
Based on a cellphone-recorded cough, machine learning models accurately detect coronavirus even in people with no symptoms.
Again and again, experts have pleaded that we need more and faster testing to control the coronavirus pandemicand many have suggested that artificial intelligence (AI) can help. Numerous COVID-19 diagnostics in development use AI to quickly analyze X-ray or CT scans, but these techniques require a chest scan at a medical facility.
Since the spring, research teams have been working toward anytime, anywhere apps that could detect coronavirus in the bark of a cough. In June, a team at the University of Oklahoma showed it was possible to distinguish a COVID-19 cough from coughs due to other infections, and now a paper out of MIT, using the largest cough dataset yet, identifies asymptomatic people with a remarkable 100 percentdetection rate.
If approved by the FDA and other regulators, COVID-19cough apps, in which a person records themselves coughing on command,could eventually be used for free, large-scale screening of the population.
With potential like that, the field is rapidly growing: Teams pursuing similar projects include a Bill and Melinda Gates Foundation-funded initiative, Cough Against Covid, at the Wadhwani Institute for Artificial Intelligence in Mumbai; the Coughvid project out of the Embedded Systems Laboratory of the cole Polytechnique Fdrale de Lausanne in Switzerland; and the University of Cambridges COVID-19 Sounds project.
The fact that multiple models can detect COVID in a cough suggeststhat there is no such thing astruly asymptomatic coronavirus infectionphysical changes alwaysoccurthat change the way a person produces sound. There arent many conditions that dont give you any symptoms, says Brian Subirana, director of the MIT Auto-ID lab and co-author on the recent study, published in the IEEE Open Journal of Engineering in Medicine and Biology.
To continue reading this article, click here.
See the original post here:
AI Recognizes COVID-19 in the Sound of a Cough Machine Learning Times - The Predictive Analytics Times