A recurrent neural network that infers the global temporal structure based on local examples
"Every day, we manipulate information about the world to make predictions,"
Jason Kim, one of the researchers who carried out the study, told TechXplore.
"How much longer can I cook this pasta before it becomes soggy? How much later
can I leave for work before rush hour? Such information representation and
computation broadly fall into the category of working memory. While we can
program a computer to build models of pasta texture or commute times, our
primary objective was to understand how a neural network learns to build models
and make predictions only by observing examples." Kim, his mentor Danielle S.
Bassett and the rest of their team showed that the two key mechanisms through
which a neural network learns to make predictions are associations and context.
For instance, if they wanted to teach their RNN to change the pitch of a song,
they fed it the original song and two other versions of it, one with a slightly
higher pitch and the other with a slightly lower pitch. For each shift in pitch,
the researchers 'biased' the RNN with a context variable. Subsequently, they
trained it to store the original and modified songs within its working
memory.Cybersecurity industry analysis: Another recurring vulnerability we must correct
Security tooling is a must-have, but we need to look wider and restore balance to the people component of security defense. Automation is the future. Why should we care about the human element of cybersecurity? Virtually everything in our lives is powered by software, and it’s true that automation is replacing the human elements that were once present in so many industries. It’s a sign of progress in a world digitizing at warp speed, with AI and machine learning hot topics keeping many organizations future-focused. So, why, then, would a human-focused approach to cybersecurity be anything other than an antiquated solution to a technologically advancing problem? The fact that billions of data records have been stolen in breaches in the past year, including the most recent Facebook breach affecting over half a billion accounts, should indicate that we’re not doing enough (or taking the right approach) to make a serious counter-punch against threat actors. Cybersecurity tooling is a much-needed component of cyber defense, and tools will always have a place. Analysts have been absolutely on point in recommending the latest tools in a risk mitigation approach for enterprises, and that will not change.Researchers Confront Major Hurdle in Quantum Computing
A time crystal is a strange state of matter in which interactions between the
particles that make up the crystal can stabilize oscillations of the system in
time indefinitely. Imagine a clock that keeps ticking forever; the pendulum of
the clock oscillates in time, much like the oscillating time crystal. By
implementing a series of electric-field pulses on electrons, the researchers
were able to create a state similar to a time crystal. They found that they
could then exploit this state to improve the transfer of an electron’s spin
state in a chain of semiconductor quantum dots. “Our work takes the first steps
toward showing how strange and exotic states of matter, like time crystals, can
potentially by used for quantum information processing applications, such as
transferring information between qubits,” Nichol says. “We also theoretically
show how this scenario can implement other single- and multi-qubit operations
that could be used to improve the performance of quantum computers.” Both AQT
and time crystals, while different, could be used simultaneously with quantum
computing systems to improve performance.How Ethical Hackers Play An Important Role In Protecting Enterprise Data
Data is an essential asset in the current dynamic setting. The value of data has made big organizations more vulnerable to cyberattacks. But believing that a big company can only suffer from a data breach incident is wrong. In reality, No one is immune to data theft, whether you’re an individual, an SME, a large enterprise, or even a state. A surer way by which organizations can protect themselves from the possibility of an effective malicious attack is to engage with competent, ethical hackers. It would help if your organization structure had someone who understands how malicious hackers think. In such scenarios, it makes sense to take the help of ethical hackers. Ethical hacking in cybersecurity has its groundwork on data protection. Unlike cybercriminals, ethical hackers operate with the consent of the client. They use the same tools and techniques as malicious attackers. However, cybersecurity and ethical hacking experts intend to protect and secure your network as they can think like the bad guys. They can quickly discover your system vulnerabilities and suggest how you can resolve them before they are exploited.Microsoft, GPT-3, and the future of OpenAI
There’s a clear line between academic research and commercial product
development. In academic AI research, the goal is to push the boundaries of
science. This is exactly what GPT-3 did. OpenAI’s researchers showed that with
enough parameters and training data, a single deep learning model could perform
several tasks without the need for retraining. And they have tested the model on
several popular natural language processing benchmarks. But in commercial
product development, you’re not running against benchmarks such as GLUE and
SQuAD. You must solve a specific problem, solve it ten times better than the
incumbents, and be able to run it at scale and in a cost-effective manner.
Therefore, if you have a large and expensive deep learning model that can
perform ten different tasks at 90 percent accuracy, it’s a great scientific
achievement. But when there are already ten lighter neural networks that perform
each of those tasks at 99 percent accuracy and a fraction of the price, then
your jack-of-all-trades model will not be able to compete in a profit-driven
market.Has DevOps killed the BA/QA/DBA Roles?
As the industry continues towards DevOps and Cloud, these fields will thin
out. Each of the roles will trend towards more of a specialization, especially
the DBA, since the operational overhead of maintaining a DB is rapidly
decreasing. They’ll last longer at big companies, but the tolerance for lower
performers will drastically decline. However, simultaneously the demand for
data expertise will keep accelerating as shown in the forecast below. Growth
in warehousing and data science should ensure data specialization remains
lucrative, and DBAs are well-poised to transition. Of the three, the BA role
seems safest. The average software developer simply does not have the time
(nor often capabilities) to maintain the social network of a strong BA.
However, as more companies migrate to DevOps/Agile, the feedback barrier
between users and developers will continue to shrink. As it does, BAs that are
not technically competent will be pushed out. The QA role is the hardest to
predict. As automation improves, demand for QA persons to run manual scripts
and “catch bugs” will disappear.
There are a bunch of certifications, from CompTIA’s Security+ to others that
will help signal your readiness for cybersecurity jobs. Some are more
entry-level and require IT competencies such as the A+. But some will require
you to have job experience in cybersecurity (such as the CISSP). There’s a bit
of a chicken and egg situation and you might wonder – how can you get job
experience if you need job experience to get the job in the first place?
Adjacent job experience can often make a difference here. Many people
transition into cybersecurity from IT roles, such as network administration,
system administration, or being on helpdesk for IT, which is an entry-level
role. You can gain experience here and transition over. There are also
programs tailored for veterans and people with law enforcement backgrounds to
get into cybersecurity. Lastly there are many cybersecurity internships being
offered to bridge this gap – though with the right backing, training, and the
right experience, you can skip ahead to junior-level analyst roles. SOC
analyst roles are a good way to break into the cybersecurity industry.
Security operations centers need analysts to parse through different
threats.
The first benefit of serverless machine learning is that it is very scalable.
It can stack up to 10k requests at the same time without having to write any
additional logic. It doesn’t consume extra time to scale which makes it
perfect for handling random high loads. Secondly, with a pay-as-you-go
architecture of serverless machine learning a person doesn’t have to pay
unused server time. It can save an enormous amount of money. For example, if a
user has 50k requests a month, he is obliged to pay only for 50k requests.
Thirdly, infrastructure management becomes very easy as a user doesn’t have to
hire a special person to look into it, it can be done very easily by a backend
developer. For instance, AWS Lambda is one of the most popular serverless
cloud services that has these advantages. It lets users run code without
managing servers. It obviated the need for developers to explicitly configure,
deploy, and manage long-term computing units. Training in Serverless Machine
Learning does not require extensive programming knowledge. Basic knowledge of
Python, Machine Learning, Linux, and Terminal along with an AWS account is
enough to get one started.
The distributed aspect of blockchain means that data are replicated across
several computers. This fact makes the hacking more challenging since there
are now several target devices. The redundancy in storage brought by
blockchain technology brings extra security and enhances data access since
users in IoT ecosystems can submit to and retrieve their data from different
devices, Carvahlo said. Continuing with this example, say the burglar is
captured and claims in court that the recorded video is forged evidence. The
immutability nature of blockchain technology means that any change to the
stored data can be easily detected. Thus, the burglar’s claim can be verified
by looking at attempts to tamper with the data, he said. However, the
decentralization aspect of blockchain technology can be a major issue when
storing data from IoT devices, according to Carvahlo. “Decentralization means
that the computers used to store data [in a distributed fashion] might belong
to different entities,” he said. “In other words, if not implemented
appropriately, there is a risk that users’ sensitive data can now be by
default stored by and available to third parties.”
How to Get a Cybersecurity Job in 2021
There are a bunch of certifications, from CompTIA’s Security+ to others that
will help signal your readiness for cybersecurity jobs. Some are more
entry-level and require IT competencies such as the A+. But some will require
you to have job experience in cybersecurity (such as the CISSP). There’s a bit
of a chicken and egg situation and you might wonder – how can you get job
experience if you need job experience to get the job in the first place?
Adjacent job experience can often make a difference here. Many people
transition into cybersecurity from IT roles, such as network administration,
system administration, or being on helpdesk for IT, which is an entry-level
role. You can gain experience here and transition over. There are also
programs tailored for veterans and people with law enforcement backgrounds to
get into cybersecurity. Lastly there are many cybersecurity internships being
offered to bridge this gap – though with the right backing, training, and the
right experience, you can skip ahead to junior-level analyst roles. SOC
analyst roles are a good way to break into the cybersecurity industry.
Security operations centers need analysts to parse through different
threats.Making A Case For Serverless Machine Learning
The first benefit of serverless machine learning is that it is very scalable.
It can stack up to 10k requests at the same time without having to write any
additional logic. It doesn’t consume extra time to scale which makes it
perfect for handling random high loads. Secondly, with a pay-as-you-go
architecture of serverless machine learning a person doesn’t have to pay
unused server time. It can save an enormous amount of money. For example, if a
user has 50k requests a month, he is obliged to pay only for 50k requests.
Thirdly, infrastructure management becomes very easy as a user doesn’t have to
hire a special person to look into it, it can be done very easily by a backend
developer. For instance, AWS Lambda is one of the most popular serverless
cloud services that has these advantages. It lets users run code without
managing servers. It obviated the need for developers to explicitly configure,
deploy, and manage long-term computing units. Training in Serverless Machine
Learning does not require extensive programming knowledge. Basic knowledge of
Python, Machine Learning, Linux, and Terminal along with an AWS account is
enough to get one started.How Blockchain Technology Can Benefit the Internet of Things
The distributed aspect of blockchain means that data are replicated across
several computers. This fact makes the hacking more challenging since there
are now several target devices. The redundancy in storage brought by
blockchain technology brings extra security and enhances data access since
users in IoT ecosystems can submit to and retrieve their data from different
devices, Carvahlo said. Continuing with this example, say the burglar is
captured and claims in court that the recorded video is forged evidence. The
immutability nature of blockchain technology means that any change to the
stored data can be easily detected. Thus, the burglar’s claim can be verified
by looking at attempts to tamper with the data, he said. However, the
decentralization aspect of blockchain technology can be a major issue when
storing data from IoT devices, according to Carvahlo. “Decentralization means
that the computers used to store data [in a distributed fashion] might belong
to different entities,” he said. “In other words, if not implemented
appropriately, there is a risk that users’ sensitive data can now be by
default stored by and available to third parties.”Software Engineering at Google: Practices, Tools, Values, and Culture
The skills required for developing good software are not the same skills that
were required (at one point) to mass produce automobiles, etc. We need
engineers to respond creatively, and to continually learn, not do one thing
over and over. If they don’t have creative freedom, they will not be able to
evolve with the industry as it, too, rapidly changes. To foster that
creativity, we have to allow people to be human, and to foster a team climate
of trust, humility, and respect. Trust to do the right thing. Humility to
realize you can’t do it alone and can make mistakes. ... Building with a
diverse team is, in our belief, critical to making sure that the needs of a
more diverse user base are met. We see that historically: first-generation
airbags were terribly dangerous for anyone that wasn’t built like the people
on the engineering teams designing those safety systems. Crash test dummies
were built for the average man, and the results were bad for women and
children, for instance. In other words, we’re not just working to build for
everyone, we’re working to build with everyone. It takes a lot of
institutional support and local energy to really build multicultural capacity
in an organization. We need allies, training, and support structures.
Quote for the day:
"Nothing so conclusively proves a man's ability to lead others as what he does from day to day to lead himself." -- Thomas J. Watson
Quote for the day:
"Nothing so conclusively proves a man's ability to lead others as what he does from day to day to lead himself." -- Thomas J. Watson
