Code Embedding: A Comprehensive Guide

Code embeddings are a transformative way to represent code snippets as dense vectors in a continuous space. These embeddings capture the semantic and functional relationships between code snippets, enabling powerful applications in AI-assisted programming. Similar to word embeddings in natural language processing (NLP), code embeddings position…

Local Generative AI: Shaping the Future of Intelligent Deployment

2024 is witnessing a remarkable shift in the landscape of generative AI. While cloud-based models like GPT-4 continue to evolve, running powerful generative AI directly on local devices is becoming increasingly viable and attractive. This local execution of generative AI can transform how small businesses, developers,…

What is language for?

What is language for?

Language is a defining feature of humanity, and for centuries, philosophers and scientists have contemplated its true purpose. We use language to share information and exchange ideas — but is it more than that? Do we use language not just to communicate, but to think?

In the June 19 issue of the journal Nature, McGovern Institute for Brain Research neuroscientist Evelina Fedorenko and colleagues argue that we do not. Language, they say, is primarily a tool for communication.

Fedorenko acknowledges that there is an intuitive link between language and thought. Many people experience an inner voice that seems to narrate their own thoughts. And it’s not unreasonable to expect that well-spoken, articulate individuals are also clear thinkers. But as compelling as these associations can be, they are not evidence that we actually use language to think.

“I think there are a few strands of intuition and confusions that have led people to believe very strongly that language is the medium of thought,” she says. “But when they are pulled apart thread by thread, they don’t really hold up to empirical scrutiny.”

Separating language and thought

For centuries, language’s potential role in facilitating thinking was nearly impossible to evaluate scientifically. But neuroscientists and cognitive scientists now have tools that enable a more rigorous consideration of the idea. Evidence from both fields, which Fedorenko, MIT brain and cognitive scientist and linguist Edward Gibson, and University of California at Berkeley cognitive scientist Steven Piantadosi review in their Nature Perspective, supports the idea that language is a tool for communication, not for thought.

“What we’ve learned by using methods that actually tell us about the engagement of the linguistic processing mechanisms is that those mechanisms are not really engaged when we think,” Fedorenko says. Also, she adds, “you can take those mechanisms away, and it seems that thinking can go on just fine.”

Over the past 20 years, Fedorenko and other neuroscientists have advanced our understanding of what happens in the brain as it generates and understands language. Now, using functional MRI to find parts of the brain that are specifically engaged when someone reads or listens to sentences or passages, they can reliably identify an individual’s language-processing network. Then they can monitor those brain regions while the person performs other tasks, from solving a sudoku puzzle to reasoning about other people’s beliefs.

“Pretty much everything we’ve tested so far, we don’t see any evidence of the engagement of the language mechanisms,” Fedorenko says. “Your language system is basically silent when you do all sorts of thinking.”

That’s consistent with observations from people who have lost the ability to process language due to an injury or stroke. Severely affected patients can be completely unable to process words, yet this does not interfere with their ability to solve math problems, play chess, or plan for future events. “They can do all the things that they could do before their injury. They just can’t take those mental representations and convert them into a format which would allow them to talk about them with others,” Fedorenko says. “If language gives us the core representations that we use for reasoning, then … destroying the language system should lead to problems in thinking as well, and it really doesn’t.”

Conversely, intellectual impairments do not always associate with language impairment; people with intellectual disability disorders or neuropsychiatric disorders that limit their ability to think and reason do not necessarily have problems with basic linguistic functions. Just as language does not appear to be necessary for thought, Fedorenko and colleagues conclude that it is also not sufficient to produce clear thinking.

Language optimization

In addition to arguing that language is unlikely to be used for thinking, the scientists considered its suitability as a communication tool, drawing on findings from linguistic analyses. Analyses across dozens of diverse languages, both spoken and signed, have found recurring features that make them easy to produce and understand. “It turns out that pretty much any property you look at, you can find evidence that languages are optimized in a way that makes information transfer as efficient as possible,” Fedorenko says.

That’s not a new idea, but it has held up as linguists analyze larger corpora across more diverse sets of languages, which has become possible in recent years as the field has assembled corpora that are annotated for various linguistic features. Such studies find that across languages, sounds and words tend to be pieced together in ways that minimize effort for the language producer without muddling the message. For example, commonly used words tend to be short, while words whose meanings depend on one another tend to cluster close together in sentences. Likewise, linguists have noted features that help languages convey meaning despite potential “signal distortions,” whether due to attention lapses or ambient noise.

“All of these features seem to suggest that the forms of languages are optimized to make communication easier,” Fedorenko says, pointing out that such features would be irrelevant if language was primarily a tool for internal thought.

“Given that languages have all these properties, it’s likely that we use language for communication,” she says. She and her coauthors conclude that as a powerful tool for transmitting knowledge, language reflects the sophistication of human cognition — but does not give rise to it. 

Studying astrophysically relevant plasma physics

Studying astrophysically relevant plasma physics

Thomas Varnish loves his hobbies — knitting, baking, pottery — it’s a long list. His latest interest is analog film photography. A picture with his mother and another with his boyfriend are just a few of Varnish’s favorites. “These moments of human connection are the ones I like,” he says.

Varnish’s love of capturing a fleeting moment on film translates to his research when he conducts laser interferometry on plasmas using off-the-shelf cameras. At the Department of Nuclear Science and Engineering, the third-year doctoral student studies various facets of astrophysically relevant fundamental plasma physics under the supervision of Professor Jack Hare.

It’s an area of research that Varnish arrived at organically.

A childhood fueled by science

Growing up in Warwickshire, England, Varnish fell in love with lab experiments as a middle-schooler after joining the science club. He remembers graduating from the classic egg-drop experiment to tracking the trajectory of a catapult, and eventually building his own model electromagnetic launch system. It was a set of electromagnets and sensors spaced along a straight track that could accelerate magnets and shoot them out the end. Varnish demonstrated the system by using it to pop balloons. Later, in high school, being a part of the robotics club team got him building a team of robots to compete in RoboCup, an international robot soccer competition. Varnish also joined the astronomy club, which helped seed an interest in the adjacent field of astrophysics.

Varnish moved on to Imperial College London to study physics as an undergraduate but he was still shopping around for definitive research interests. Always a hands-on science student, Varnish decided to give astronomy instrumentation a whirl during a summer school session in Canada.

However, even this discipline didn’t quite seem to stick until he came upon a lab at Imperial conducting research in experimental astrophysics. Called MAGPIE (The Mega Ampere Generator for Plasma Implosion Experiments), the facility merged two of Varnish’s greatest loves: hands-on experiments and astrophysics. Varnish eventually completed an undergraduate research opportunity (UROP) project at MAGPIE under the guidance of Hare, his current advisor, who was then a postdoc at the MAGPIE lab at Imperial College.

Part of Varnish’s research for his master’s degree at Imperial involved stitching together observations from the retired Herschel Space Telescope to create the deepest far-infrared image ever made by the instrument. The research also used statistical techniques to understand the patterns of brightness distribution in the images and to trace them to specific combinations of galaxy occurrences. By studying patterns in the brightness of a patch of dark sky, Varnish could discern the population of galaxies in the region.

Move to MIT

Varnish followed Hare (and a dream of studying astrophysics) to MIT, where he primarily focuses on plasma in the context of astrophysical environments. He studies experimental pulsed-power-driven magnetic reconnection in the presence of a guide field.

Key to Varnish’s experiments is a pulsed-power facility, which is essentially a large capacitor capable of releasing a significant surge of current. The electricity passes through (and vaporizes) thin wires in a vacuum chamber to create a plasma. At MIT, the facility currently being built at the Plasma Science and Fusion Center (PSFC) by Hare’s group is called: PUFFIN (PUlser For Fundamental (Plasma Physics) INvestigations).

In a pulsed-power facility, tiny cylindrical arrays of extremely thin metal wires usually generate the plasma. Varnish’s experiments use an array in which graphite leads, the kind used in mechanical pencils, replace the wires. “Doing so gives us the right kind of plasma with the right kind of properties we’d like to study,” Varnish says. The solution is also easy to work with and “not as fiddly as some other methods.” A thicker post in the middle completes the array. A pulsed current traveling down the array vaporizes the thin wires into a plasma. The interactions between the current flowing through the plasma and the generated magnetic field pushes the plasma radially outward. “Each little array is like a little exploding bubble of magnetized plasma,” Varnish says. He studies the interaction between the plasma flows at the center of two adjacent arrays.

Studying plasma behavior

The plasma generated in these pulsed-power experiments is stable only for a few hundred nanoseconds, so diagnostics have to take advantage of an extremely short sampling window. Laser interferometry, which images plasma density, is Varnish’s favorite. In this technique, a camera takes a picture of a split laser beam, one arm of which encounters the plasma and one that doesn’t. The arm that hits the plasma produces an interference pattern when the two arms are recombined. Capturing the result with a camera allows researchers to infer the structure of the plasma flows.

Another diagnostic method involves placing tiny loops of metal wire in the plasma (called B-dots), which record how the magnetic field in the plasma changes in time. Yet another way to study plasma physics is using a technique called Faraday rotation, which measures the twisting of polarized light as it passes through a magnetic field. The net result is an “image map of magnetic fields, which is really quite incredible,” Varnish says.

These diagnostic techniques help Varnish research magnetic reconnection, the process by which plasma breaks and reforms magnetic fields. It’s all about energy redistribution, Varnish says, and is particularly relevant because it creates solar flares. Varnish studies how having not-perfectly-opposite magnetic field lines might affect the reconnection process.

Most research in plasma physics can be neatly explained by the principles of magnetohydrodynamics, but the phenomena observed in Varnish’s experiments need to be explained with additional theories. Using pulsed power enables studies over longer length scales and time periods than in other experiments, such as laser-driven ones. Varnish is looking forward to working on simulations and follow-up experiments on PUFFIN to study these phenomena under slightly different conditions, which might shed new light on the processes.

At the moment, Varnish’s focus is on programming the control systems for PUFFIN so he can get it up and running. Part of the diagnostics system involves ensuring that the facility will deliver the plasma-inducing currents needed and perform as expected.

Aiding LGBTQ+ efforts

When not working on PUFFIN or his experiments, Varnish serves as co-lead of an LGBTQ+ affinity group at the PSFC, which he set up with a fellow doctoral student. The group offers a safe space for LGBTQ+ scientists and meets for lunch about once a month. “It’s been a nice bit of community building, and I think it’s important to support other LGBTQ+ scientists and make everyone feel welcome, even if it’s just in small ways,” Varnish says, “It has definitely helped me to feel more comfortable knowing there’s a handful of fellow LGBTQ+ scientists at the center.”

Varnish has his hobbies going. One of his go-to bakes is a “rocky road,” a British chocolate bar that mixes chocolate, marshmallows, and graham crackers. His research interests, too, are a delicious concoction mixed together: “the intersection of plasma physics, laboratory astrophysics, astrophysics (the won’t-fit-in-a-lab kind), and instrumentation.”

Apple secures ‘observer’ seat on OpenAI board

Following Apple’s partnership announcement with OpenAI at WWDC last month, a new report reveals that the tech giant will secure an “observer role” on OpenAI’s board of directors. The new arrangement – set to take effect later this year – will see Apple’s long-time marketing chief…

Signal processing: How did we get to where we’re going?

On May 24, Ford Professor of Engineering Al Oppenheim addressed a standing-room-only audience at MIT to give the talk of a lifetime. Entitled “Signal Processing: How Did We Get to Where We’re Going?”, Oppenheim’s personal account of his involvement in the early years of the digital signal processing field included a photo retrospective — and some handheld historical artifacts — that showed just how far the field has come since its birth at MIT and Lincoln Laboratory. Hosted by Anantha Chandrakasan, chief innovation and strategy officer, dean of engineering, and the Vannevar Bush Professor of Electrical Engineering and Computer Science, the event included a lively Q & A, giving students the chance to gain Oppenheim’s insight about the trajectory of this ever-growing field.

Signal processing: How did we get to where we’re going?

Play video

Al Oppenheim: “Signal Processing: How did we get to where we’re going?”

Al Oppenheim received a ScD degree in 1964 at MIT and is also the recipient of an honorary doctorate from Tel Aviv University. During his career, he has been a member of the Research Laboratory of Electronics and closely affiliated with MIT Lincoln Laboratory and with the Woods Hole Oceanographic Institution. His research interests are in the general area of signal processing algorithms, systems, and applications. He is co-author of the widely used textbooks “Digital Signal Processing,” “Discrete-Time Signal Processing” (currently in its third edition), “Signals and Systems” (currently in its second edition), and most recently “Signals, Systems & Interference,” published in 2016. He is also the author of several video courses available online. He is editor of several advanced books on signal processing. Throughout his career he has published extensively in research journals and conference proceedings.

Oppenheim is a member of the National Academy of Engineering, an IEEE Life Fellow, and has been a  Guggenheim Fellow in France and a Sackler Fellow in Israel. He has received a number of IEEE awards for outstanding research, teaching, and mentoring, including the IEEE Kilby Medal; the IEEE Education Medal; the IEEE Centennial Award; the IEEE Third Millennium Medal; the Norbert Wiener Society award; and the Society, Technical Achievement, and Senior Awards of the IEEE Society on Acoustics, Speech and Signal Processing; as well as a number of research, teaching, and mentoring awards at MIT.

 

Tales Of Kenzera: Zau Developer Surgent Studios Announces Layoffs

Tales Of Kenzera: Zau Developer Surgent Studios Announces Layoffs

Surgent Studios, the developer behind Tales of Kenzera: Zau, has announced it is laying off just over a dozen employees. This follows the release of its first game, Tales of Kenzera: Zau, back in April, which launched to good reviews

Surgent doesn’t explain why; it says it’s focusing on supporting those affected, continuing work on Tales of Kenzera: Zau, and looking ahead to future projects. 

Here’s the studio’s statement, in full

“Unfortunately, Surgent has joined the growing number of games studios impacted by layoffs this year with just over a dozen people affected. It’s a difficult time in the games industry, but we remain incredibly proud of our entire team’s work on Zau, and of the praise it has received from critics and players alike. Our focus now is on supporting those affected, continuing our work on Zau, and looking to the future with our next creative projects.” 

Studio head and founder Abubakar Salim, who also voiced the titular Zau in the team’s first game, released his own statement addressing the layoffs: 

“Thank you so much to those who have checked in. This hurts deeply. This isn’t the news I wanted to share today. I am so proud of what that team have achieved over the course of these 4 years. When things got tough, every one of them stood so strong, it was inspiring. So to be delivering this news today really sucks. I know we’re not alone here, but that doesn’t make it easier. 

“The focus now is to continue supporting those affected in anyway we can. I will be replying to this with links to posts from our affected team members. If you have any opportunities available or know of any going, please consider these incredible talented people.” 

Surgent Studios grows an unfortunately ever-growing list of studios affected by layoffs in 2024. 


Last month, Paradox Interactive closed its Paradox Tectonic Studio the same week it canceled its first game, Life by You. Earlier that month, Dead by Daylight developer Behaviour Interactive laid off 95 employees. Fae Farm and Dauntless developer Phoenix Labs laid off the majority of its staff and canceled its in-development games back in May, and that same week, Square Enix announced it will begin layoffs as part of “structural reforms.” 

In May, Xbox closed four Bethesda studios, including Hi-Fi Rush developer Tango Gameworks and Redfall studio Arkane Austin. Take-Two Interactive closed Rollerdrome studio Roll7 and Kerbal Space Program 2 studio Intercept Games alongside major layoffs to its indie-publisher Private Division label. That same week, we learned Deliver Us Mars developer Keoken Interactive had laid off nearly its entire staff

Elsewhere in the year, EA laid off roughly 670 employees across all departments, resulting in the cancellation of Respawn’s Star Wars FPS game. PlayStation laid off 900 employees across Insomniac, Naughty Dog, Guerrilla, and more, closing down London Studio in the process, too. The day before, Until Dawn developer Supermassive Games announced it laid off 90 employees

At the end of January, we learned Embracer Group had canceled a new Deus Ex game in development at Eidos-Montréal and laid off 97 employees in the process. Also in January, Destroy All Humans remake developer Black Forest Games reportedly laid off 50 employees and Microsoft announced it was laying off 1,900 employees across its Xbox, Activision Blizzard, and ZeniMax teams, as well. Outriders studio People Can Fly laid off more than 30 employees in January, and League of Legends company Riot Games laid off 530 employees

Lords of the Fallen Publisher CI Games laid off 10 percent of its staffUnity will lay off 1,800 people by the end of March, and Twitch laid off 500 employees

We also learned that Discord had laid off 170 employees, that layoffs happened at PTW, a support studio that’s worked with companies like Blizzard and Capcom, and that SteamWorld Build company, Thunderful Group, let go of roughly 100 people. Dead by Daylight developer Behaviour Interactive also reportedly laid off 45 people, too

The hearts of the Game Informer staff are with everyone who’s been affected by layoffs or closures. 

Google’s dilemma: AI expansion vs achieving climate goals

Google is currently grappling with a major dilemma: finding a way to maintain its impressive progress in AI technology while also staying true to its goal of minimising carbon emissions. In its 2024 Environmental Report, Google discloses a concerning trend: a staggering 50% surge in emissions…

How AI is transforming digital marketing: 2024 trends and insights – AI News

Discover how AI is revolutionising digital marketing with success stories and key strategies. Learn about personalisation, predictive analytics, content creation, and more. The rapid evolution of AI is revolutionising digital marketing, offering unprecedented opportunities for personalisation, efficiency, and customer engagement. By leveraging advanced algorithms and machine…