Optimizing Memory for Large Language Model Inference and Fine-Tuning

Large language models (LLMs) like GPT-4, Bloom, and LLaMA have achieved remarkable capabilities by scaling up to billions of parameters. However, deploying these massive models for inference or fine-tuning is challenging due to their immense memory requirements. In this technical blog, we will explore techniques for…

Epigenomic analysis sheds light on risk factors for ALS

Epigenomic analysis sheds light on risk factors for ALS

For most patients, it’s unknown exactly what causes amyotrophic lateral sclerosis (ALS), a disease characterized by degeneration of motor neurons that impairs muscle control and eventually leads to death.

Studies have identified certain genes that confer a higher risk of the disease, but scientists believe there are many more genetic risk factors that have yet to be discovered. One reason why these drivers have been hard to find is that some are found in very few patients, making it hard to pick them out without a very large sample of patients. Additionally, some of the risk may be driven by epigenomic factors, rather than mutations in protein-coding genes.

Working with the Answer ALS consortium, a team of MIT researchers has analyzed epigenetic modifications — tags that determine which genes are turned on in a cell — in motor neurons derived from induced pluripotent stem (IPS) cells from 380 ALS patients.

This analysis revealed a strong differential signal associated with a known subtype of ALS, and about 30 locations with modifications that appear to be linked to rates of disease progression in ALS patients. The findings may help scientists develop new treatments that are targeted to patients with certain genetic risk factors.

“If the root causes are different for all these different versions of the disease, the drugs will be very different and the signals in IPS cells will be very different,” says Ernest Fraenkel, the Grover M. Hermann Professor in Health Sciences and Technology in MIT’s Department of Biological Engineering and the senior author of the study. “We may get to a point in a decade or so where we don’t even think of ALS as one disease, where there are drugs that are treating specific types of ALS that only work for one group of patients and not for another.”

MIT postdoc Stanislav Tsitkov is the lead author of the paper, which appears today in Nature Communications.

Finding risk factors

ALS is a rare disease that is estimated to affect about 30,000 people in the United States. One of the challenges in studying the disease is that while genetic variants are believed to account for about 50 percent of ALS risk (with environmental factors making up the rest), most of the variants that contribute to that risk have not been identified.

Similar to Alzheimer’s disease, there may be a large number of genetic variants that can confer risk, but each individual patient may carry only a small number of those. This makes it difficult to identify the risk factors unless scientists have a very large population of patients to analyze.

“Because we expect the disease to be heterogeneous, you need to have large numbers of patients before you can pick up on signals like this. To really be able to classify the subtypes of disease, we’re going to need to look at a lot of people,” Fraenkel says.

About 10 years ago, the Answer ALS consortium began to collect large numbers of patient samples, which could allow for larger-scale studies that might reveal some of the genetic drivers of the disease. From blood samples, researchers can create induced pluripotent stem cells and then induce them to differentiate into motor neurons, the cells most affected by ALS.

“We don’t think all ALS patients are going to be the same, just like all cancers are not the same. And the goal is being able to find drivers of the disease that could be therapeutic targets,” Fraenkel says.

In this study, Fraenkel and his colleagues wanted to see if patient-derived cells could offer any information about molecular differences that are relevant to ALS. They focused on epigenomic modifications, using a method called ATAC-seq to measure chromatin density across the genome of each cell. Chromatin is a complex of DNA and proteins that determines which genes are accessible to be transcribed by the cell, depending on how densely packed the chromatin is.

In data that were collected and analyzed over several years, the researchers did not find any global signal that clearly differentiated the 380 ALS patients in their study from 80 healthy control subjects. However, they did find a strong differential signal associated with a subtype of ALS, characterized by a genetic mutation in the C9orf72 gene.

Additionally, they identified about 30 regions that were associated with slower rates of disease progression in ALS patients. Many of these regions are located near genes related to the cellular inflammatory response; interestingly, several of the identified genes have also been implicated in other neurodegenerative diseases, such as Parkinson’s disease.

“You can use a small number of these epigenomic regions and look at the intensity of the signal there, and predict how quickly someone’s disease will progress. That really validates the hypothesis that the epigenomics can be used as a filter to better understand the contribution of the person’s genome,” Fraenkel says.

“By harnessing the very large number of participant samples and extensive data collected by the Answer ALS Consortium, these studies were able to rigorously test whether the observed changes might be artifacts related to the techniques of sample collection, storage, processing, and analysis, or truly reflective of important biology,” says Lyle Ostrow, an associate professor of neurology at the Lewis Katz School of Medicine at Temple University, who was not involved in the study. “They developed standard ways to control for these variables, to make sure the results can be accurately compared. Such studies are incredibly important for accelerating ALS therapy development, as they will enable data and samples collected from different studies to be analyzed together.”

Targeted drugs

The researchers now hope to further investigate these genomic regions and see how they might drive different aspects of ALS progression in different subsets of patients. This could help scientists develop drugs that might work in different groups of patients, and help them identify which patients should be chosen for clinical trials of those drugs, based on genetic or epigenetic markers.

Last year, the U.S. Food and Drug Administration approved a drug called tofersen, which can be used in ALS patients with a mutation in a gene called SOD1. This drug is very effective for those patients, who make up about 1 percent of the total population of people with ALS. Fraenkel’s hope is that more drugs can be developed for, and tested in, people with other genetic drivers of ALS.

“If you had a drug like tofersen that works for 1 percent of patients and you just gave it to a typical phase two clinical trial, you probably wouldn’t have anybody with that mutation in the trial, and it would’ve failed. And so that drug, which is a lifesaver for people, would never have gotten through,” Fraenkel says.

The MIT team is now using an approach called quantitative trait locus (QTL) analysis to try to identify subgroups of ALS patients whose disease is driven by specific genomic variants.

“We can integrate the genomics, the transcriptomics, and the epigenomics, as a way to find subgroups of ALS patients who have distinct phenotypic signatures from other ALS patients and healthy controls,” Tsitkov says. “We have already found a few potential hits in that direction.”

The research was funded by the Answer ALS program, which is supported by the Robert Packard Center for ALS Research at Johns Hopkins University, Travelers Insurance, ALS Finding a Cure Foundation, Stay Strong Vs. ALS, Answer ALS Foundation, Microsoft, Caterpillar Foundation, American Airlines, Team Gleason, the U.S. National Institutes of Health, Fishman Family Foundation, Aviators Against ALS, AbbVie Foundation, Chan Zuckerberg Initiative, ALS Association, National Football League, F. Prime, M. Armstrong, Bruce Edwards Foundation, the Judith and Jean Pape Adams Charitable Foundation, Muscular Dystrophy Association, Les Turner ALS Foundation, PGA Tour, Gates Ventures, and Bari Lipp Foundation. This work was also supported, in part, by grants from the National Institutes of Health and the MIT-GSK Gertrude B. Elion Research Fellowship Program for Drug Discovery and Disease.

JVC KY-PZ540 40x NDI HX3 PTZ Camera is a Winner! – Videoguys

JVC KY-PZ540 40x NDI HX3 PTZ Camera is a Winner! – Videoguys

In a recent blog post authored by Terra York for Videomaker, JVC’s KY-PZ540 Series PTZ Cameras have been honored with the prestigious title of Best Camera of NAB 2024. This recognition underscores the remarkable capabilities of these cameras, especially their groundbreaking 40x zoom functionality—a first among PTZ cameras. The KY-PZ540 cameras are equipped with a 4K CMOS sensor and feature a 20x optical zoom for 4K streaming, seamlessly transitioning to HD with JVC’s Variable Scan Mapping (VSM) technology to deliver a full 40x zoom resolution in HD mode.

The design of the KY-PZ540 cameras incorporates intuitive elements, including a two-color tally light, internal microphones, and an IP address display on the front panel, complemented by a diverse range of connectivity options on the rear—such as HDMI, 3G-SDI, and audio ports, as well as USB-C, RS-232C, and ethernet ports. These cameras are available in both black and white variants to suit different aesthetic preferences and environmental settings.

The KY-PZ540 Series supports streaming across various resolutions, delivering up to 60p in 4K UHD and 1080p and 720p in HD. While HD-SDI outputs are capped at 1080p, HDMI can handle 4K streaming effortlessly. Notably, the cameras feature UVC (USB Video Class) and NDI HX3 capabilities in the KY-PZ540N model, facilitating seamless network sharing and enhanced accessibility of video streams.

One of the standout features is the incorporation of H.265/HEVC compression technology, enabling efficient 4K 60p streaming—particularly advantageous for live sports and events coverage, offering improved image quality at reduced bitrates compared to H.264. The KY-PZ540 cameras leverage multi-slice encoding technology to optimize performance and minimize latency in 4K 60p HEVC mode.

Another highlight is the subject auto-tracking functionality, offering five distinct settings: Standard, Area, Stage, Wide Area, and Fine Adjustment. These settings empower users to customize tracking parameters based on specific shooting scenarios, ensuring subjects remain sharply focused and precisely framed throughout.

In summary, JVC’s KY-PZ540 Series PTZ Cameras are exceptionally well-suited for large event spaces, stage performances, sports coverage, and places of worship owing to their exceptional zoom capabilities and advanced auto-tracking features. Priced starting at $1,999, these cameras represent a compelling investment for individuals seeking to elevate their PTZ systems or establish new setups, offering unmatched versatility across various streaming and connectivity requirements. This blog post illuminates the innovative features and practical benefits of JVC’s acclaimed KY-PZ540 Series, positioning it as an indispensable tool for videographers and content creators seeking unparalleled performance in the world of PTZ cameras.

Read the full article by Terra York for Videomaker HERE

BirdDog is Back!!! Introducing the BirdDog X1, X1 Ultra, and MAKI Ultr – Videoguys

In this week’s Videoguys Live episode, Gary introduces BirdDog’s exciting new X Series cameras. Join our exclusive webinar as Gary explores the cutting-edge BirdDog X1, X1 Ultra, and MAKI Ultra cameras, highlighting their compact design and exceptional performance. Discover the future of camera technology and secure your pre-order with Videoguys today! Stay ahead with BirdDog’s innovative camera lineup.

Watch the full webinar below:

[embedded content]


BirdDog X1 and X1 Ultra PTZ Cameras

  • 1080/60p
  • 20X Zoom
  • Tally Light
  • AI Auto Focus tracking
  • HDMI/USB/IP
  • NDI HX3
  • WiFi Connection
  • E-ink Label
  • Integrated NDI HX decoder

  • 4K/30p
  • 12X Zoom
  • Tally Light
  • AI Auto Focus tracking
  • HDMI/USB/IP
  • NDI HX3
  • WiFi Connection
  • E-ink Label
  • Integrated NDI HX decoder

Tally Light. It’s a halo Effect.
Professional results need professional tools. X1 and X1 Ultra sport the most visible and effective Tally system available. Up front with a large, daylight visible border, and out back with an interchangeable, illuminated numbering system. Your cast and crew will always know where to look to keep your audience engaged and your look polished.

AI Tracking. We welcome our AI overlords.
Every production could use a little help from a friend. The X1 and X1 Ultra keep your subject in focus with AI tracking technology, intelligently following your subject’s movements in real-time. Whether you’re orchestrating a live event, capturing spontaneous moments in a hybrid environment, or hunting down sports action, AI tracking helps you get well framed – without needing an octopus to run the controller.

NDI® | HX3. Enhanced Reliability on More Networks.
The X1 and X1 Ultra deliver superior network performance via NDI® | HX3 integration. You get significantly reduced network impact and excellent image quality, meaning its more friendly in more network environments. BirdDog’s NDI® | HX3 delivers quality and performance you can count on.

Integrated NDI® | HX Decoder. Let’s hook up.
The X1 and X1 Ultra are the world’s first PTZ cameras featuring an NDI® | HX* Decoder designed specifically for confidence monitoring, return feeds, or teleprompter usage. We’ve pulled all this connectivity into a single cable, radically simplifying how you interface with your equipment. Less gear. Less hassle. Whether you’re broadcasting live or delivering a CEO address, the integrated NDI HX decoder enables more brilliance with less setup. *NDI® | HX2 and NDI® | HX3 supported

Wi-Fi. Connected but not connected.
No need to use cables to capture every shot. On top of regular ethernet with PoE, X1 and X1 Ultra have built-in Wi-Fi functionality. Easily capture shots from spots where it’s hard to run cable, seamlessly integrating into any environment. Your set up just became a whole lot more flexible.

E-Ink Label. Get Inked.
The always-on e-Ink label on the X1 and X1 Ultra are game-changers. From network addresses for easy access to controls, this cool feature ensures that your set-up is clear and each camera easily identifiable even when the power is off.

Dual-Network Output
X1 and X1 Ultra combine NDIHX and SRT/RTMP protocols for seamless on-premises and remote/CDN connectivity, ensuring ultimate flexibility in both streaming set-ups and distribution.

NDI Ecosystem
X1 and X1 Ultra can be integrated into multiple software environments including TriCaster, vMix, Wirecast, ProPresenter, Epiphan Pearl, OBS, Vizrt Vectar, FOR.A, Broascast Pix, and many more

Baseband
X1 and X1 Ultra’s baseband HDMI connectivity means it can be easily included in your existing production set-up. Connect with tech like video switchers, SDI converters, HDMI products and many more.

Serial Control
Take control effortlessly with older control protocols, seamlessly integrating with your existing infrastructure including PTZ controllers and automation systems

USB Webcam
X1 and X1 Ultra feature USB-C for connecting directly to Zoom, Teams, and most apps that support a USB UVC input

Control Support
The X1 and X1 Ultra ship with their own IR remote, plus hardware control support for BirdDog PTZ keyboard, Skaarhoj, CyanView, Streamdeck, Loupedeck, and even Xbox controllers

API 2.0 Support
With support for RESTful API, there are so many ways to program your own control over X1 and X1 Ultra to create a totally integrated PTZ camera solution.

BirdDog Apps
X1 and X1 Ultra are fully compatible with the BirdDog family apps for routing, production, testing, and monitoring, remote shading, and more. This includes NDI PTZ Control (iOS), Central, and MV Pro


MAKI Ultra Box Camera

  • 4K/60p
  • 12X & 20X available
  • Compact
  • Super-fast Auto Focus
  • HDMI/SDI/USB/IP
  • NDI HX3
  • On-board buttons
BirdDog is Back!!! Introducing the BirdDog X1, X1 Ultra, and MAKI Ultr – Videoguys

Compact. Yet Mighty.
The MAKI Ultra boasts a compact footprint helping you capture those unusual and engaging camera angles with ease. With MAKI Ultra you can deliver killer POV shots that put your audience right in the game at eSports events, or attach MAKI Ultra to a band rig to capture incredible angles of your drummers or instrumentalists, be they in a band or a house of worship. Wherever the action is, MAKI Ultra ensures you never miss a beat.

Super-Fast Autofocus. Keeping Pace with Your Vision.
It’s fast AF. MAKI Ultra’s super-fast autofocus is a game-changer for dynamic content and fast-paced environments. Equipped with a ToF sensor and powered by advanced automatic focusing algorithms, MAKI Ultra ensures rapid and accurate autofocus. Whether your subjects are zipping across the frame or moving closer and further away from the camera, MAKI Ultra keeps them sharp and in focus. Every. Single. Time.

4K60. Capture Every Moment with Ultra Detail.
With its 4K60 capabilities, MAKI Ultra expands your options on what and how you capture content. From fast-moving action shots to incredible detail in low-light sets, you can get all the detail thanks to MAKI Ultra’s high frame rate and advanced image processing capabilities.

NDI® | HX3. Enhanced Performance on More Networks.
MAKI Ultra delivers superior network performance via NDI® | HX3 integration. Significantly reducing network impact while maintaining excellent image quality, meaning it’s more friendly in more network environments. BirdDog’s NDI® | HX3 surpasses certification parameters, delivering quality and performance you can count on.

On-board Buttons. Let Your Digits Take Control.
MAKI Ultra’s intuitive physical buttons means that once you’ve set it up, you can just reach in and make quick adjustments to Zoom and Focus, and access to the onscreen menu. No awkwardly placed controls or a walk back to your controller. One and done. And on to the next task.

Baseband.
MAKI Ultra’s baseband 3G-SDI and HDMI connectivity means it can be easily included in your existing production set-up. Connect with tech like video switchers, SDI converters, HDMI products, and many more.

USB Webcam.
MAKI Ultra features USB-C for connecting directly to (deep breath) Zoom, Teams, Barco Clickshare, Google Meet, Webex, GoTo Meeting, BlueJeans, Slack, and more. Safe to say, if you have an app that works with a USB Web Camera, you can supercharge it with MAKI Ultra.

Colour Matrix.
Dialling in all your cameras changes your production quality from good to great. Level up your game with included Colour Matrix controls. MAKI Ultra has individual gain control over Red, Green, Blue, Cyan, Yellow and Magenta, all with 64 levels of control for ultra precision.

Flexible Power.
MAKI Ultra can be powered in several ways via the included DC power supply or PoE. With under 15W PoE requirement it makes it not only convenient but cost effective on infrastructure owing to its low power consumption.

Advanced Features in BirdUI.
BirdUI is our web browser based gateway to your camera. It gives access to all the functions of your MAKI Ultra, within a beautiful and intuitive interface on any device, anywhere on the network. BirdUI adds features including a dashboard for monitoring the camera and light/dark modes to suit your production environment.

NDI® Mute.
This function allows for muting of the NDI ® stream on the network. Perfect for meeting spaces, public spaces, classrooms, and any time some privacy is needed.

Why Do Web Design Projects Come to a Standstill?

A web designer’s schedule can be challenging. We try to plan so that projects don’t overlap. Scrambling to launch multiple websites simultaneously isn’t ideal. But it seems like something always disrupts our carefully crafted agenda.

A stalled project is often to blame. Everything is moving along – until it isn’t. Suddenly, your eager client has gone quiet. And you’re left to wonder and wait.

It may be a temporary situation. But it can drag on for months or longer. And the delay can have a significant impact.

Not only does the situation hurt your schedule, but it can also affect your bottom line. That money you were expecting hangs in the air along with the project. It’s not a good situation for your business.

So, why do web design projects come to a standstill? Here are a few thoughts on the root causes. Along the way, we’ll offer some tips to get things moving again.

Too Much Design by Committee

Project stakeholders must work together effectively and efficiently. But the more people involved, the harder it is to collaborate. And we can’t forget about the difficulty in achieving consensus.

The result is something that web designers loathe: design by committee. It’s a situation where everyone puts their ideas on the table. The ideas often conflict with and contradict each other. A lack of leadership and a clear vision for the project cause chaos.

This internal strife can grind progress to a halt. Even worse is that web designers are caught in the middle. We listen to the gripes but feel powerless to act.

All is not lost, however. If your client’s leadership is amiss, you might be able to help. Stepping into the void and bringing people together can restart progress.

The key is to get everyone working toward the same goal. Once that happens, things can come together quickly.

Why Do Web Design Projects Come to a Standstill?

The Woes of Client-Created Content

Clients can be great at thinking big when it comes to content. They’ll have grand ideas for in-depth pages and blog posts. They envision a website that’s teeming with compelling articles.

Ideas aren’t the problem. The ability to deliver on them gets in the way. The reality tends to hit clients when it’s time to do the work.

It’s among the most common reasons for a slowdown. Ambitious clients bite off more than they can chew. They may feel overwhelmed by the task ahead of them.

The difficulty for web designers is that content is often the last piece of the puzzle. We can’t move forward without it. Thus, a beautiful new website may sit dormant while we wait.

Thankfully, it’s another area where we can pitch in. Our clients aren’t necessarily content creators. Therefore, they need someone to act as a guide.

You can walk them through the content creation process. Provide writing prompts or templates that clients can fill in. Short of that, you might also offer to produce the content yourself.

Content-related delays can go on indefinitely. Being proactive could make a positive difference.

Web design clients may struggle in putting together content.

A Client That’s Strapped for Time

Some clients have multiple responsibilities. It’s especially the case for small organizations. A decisionmaker may wear many hats. Their website is just one.

Building a website requires time to focus on the details. A client will need to review the site’s design and content. They’ll also need to coordinate with team members. There’s also the time needed for meetings and brainstorming.

It’s a challenge even during relatively quiet times. And it can slow things down immensely when something unexpected pops up.

It can be difficult for a client to ramp things back up. Regaining momentum after a long break doesn’t come naturally. Thus, they may put the project on the back burner.

You could wait for your client to restart things. But this also adds uncertainty to your schedule.

Communication might bring the project back to life. Get in touch with your client and offer assistance. Try to encourage them to delegate tasks when possible.

Everyone has limited time to get things done. Demonstrating a willingness to help may ease a client’s burden.

A client may be busy tending to other areas of their business.

Deal with Project Delays Like a Pro

Dealing with project delays is a part of being a web designer. We may pride ourselves on being timely. But client-related stoppages are out of our control.

And there are a variety of reasons for a standstill. A client may have too much on their plate. They might also struggle with managing other stakeholders. Or they had an unexpected issue with their budget. The sky’s the limit.

The challenge for designers is that we need client participation. Sure, we can perform some tasks in the background. But we can’t finish the job without their feedback and approval.

Sometimes a delay is a positive. It frees up time to work on other projects while we wait. But that’s not always the case.

Indeed, the uncertainty can hang over our heads. When a client does get around to restarting, it inevitably clashes with our schedule. We’re left trying to squeeze the project into what we’re already doing.

The best we can do is to try and keep things moving forward. Making an effort here will be appreciated by your clients. As is knowing they aren’t alone in wanting to finish the project.

Related Topics


Top

Fallout’s Weirdest Vaults And The Games That Should Go With Them

This article originally published on June 7, 2018

With the announcement of Fallout 76, fans are scrambling for clues that will shed light on the new game. A few concrete details come from the pre-existing information on Vault 76. Most vaults in Fallout were built for a specific purpose, and according to references in Fallout 3, Vault 76 supposedly held 500 people for exactly 20 years before opening and repopulating the surface.

Repopulation is important and all, but is this the most interesting route for a new Fallout? Playing a wastelander on a noble quest is something we’ve already done several times, but a game built around the events of a specific vault is an intriguing proposition. We delve into canon Fallout lore to reveal a bounty of bizarre and disturbing scenarios, any of which could make for an excellent spinoff.

Fallout’s Weirdest Vaults And The Games That Should Go With Them

Vault 12

Vault-Tec, the company who built the shelters spread across America, is a notoriously unethical corporation in the Fallout universe. In collaboration with the U.S. Government, Vault-Tec built most of its vaults around “social experiments,” testing the effects of isolation and other variables in the tightly controlled underground communities. 

Vault 12 is one of the most twisted of these experiments. Foregoing the very purpose of a fallout shelter, Vault-Tec intentionally left the door unsealed. Awash in deadly radiation, the survivors of Vault 12 were permanently transformed by Vault-Tec’s negligence.

Ghouls are familiar to anyone who’s played a Fallout game. Zombie-like in appearance, their rotting flesh is a product of exposure to massive amounts of radiation. Despite this, they’re often surprisingly empathetic characters, with thoughtful dialogue and a depth of experience taken from their horrifying life experiences.

One of the biggest hubs of ghouls is Necropolis, a city players may run across in the original Fallout. The massive population stems from, you guessed it, the residents of Vault 12.

Here’s the pitch: Survival horror on a community scale. Players try to maintain life as a resident in the vault while people get sick and decay. You decide who gets Rad-X and who has to bear the brunt of the radiation then manage the dividing populations as people turn to ghouls and lose sympathy for their fellow residents. Imagine a combination of some of the haunted vaults in Fallout 3 and the oppressive wartime decisions of This War of Mine and you’ve got Fallout: Vault 12.  

This, for a whole game.

Vault 29

One criticism of Bethesda’s Fallout games was the removal of the ability to kill children. Although sadistic, the complaint had two main pillars. Firstly, Fallout 1 and 2 had the ability; it benefited the player in no way and made virtually every NPC in the game hostile, but it was technically possible. Secondly, kids in the Fallout games have been historically … very annoying.

Enter Vault 29, population: children. Another perversion of an experiment by Vault-Tec, no one over the age of 15 was allowed into the shelter. The only confirmed survivor, a mutant named Harold, emerged from the vault in 2090 and would go on to become one of the wasteland’s most influential residents.

Here’s the pitch: A dark, violent version of Richard Linklater’s Boyhood. Harold entered the vault at age five and experiences his entire adolescence in the underground tunnels, so players would experience his life one year at a time. The early years would be chaos, both from the initial panic of being without adults, but also because we’d experience them as a kindergartner.

As time wears on, we’d mature and start to learn about the political structure of the vault. I’m picturing Battle School from Ender’s Game with multiple factions with different leaders and styles, all led by charismatic children. Like Ender’s Game, this game would still have truly bleak themes and conflict without resorting to kids toting firearms. As Harold, we’d come to understand the dynamic of the different factions, and fight to become a part of the most influential groups.

This, but ya know. Radioactive.

Vault 43

Twenty Men. Ten Women. One Panther. That’s the idea for Vault 43, a shelter that really calls into question Vault-Tec’s understanding of the word “experiment.” Nothing is known about the outcome of this vault, and that might be for the best.

Here’s the pitch: Asymmetrical multiplayer. The men and women work together to try and build a society that’s democratic, charitable, and most importantly, panther-proof. Meanwhile, the panther is slinking in and out of the shadows, testing the defenses and eating anyone who strays too far from the beaten path.

The game’s pace could play out similarly to Evolve. In the early rounds, the panther is weak but the residents don’t have access to many resources. As time goes on, the men and women can fortify doors and board up air vents, but they won’t be able to stop the panther from accessing more radioactivity than Dr. Manhattan. Sure, the humans have numbers on their side, but how many people is enough to stop a radioactive, mutant panther?

This, but it breaks. A lot.

Vault 53

Intended to test the inhabitants’ resilience to everyday inconveniences, appliances in Vault 53 were designed to break every few months. With 1000 residents, this meant a near-constant stream of necessary repairs to maintain basic living conditions. One can only imagine the psychological toll of living inside deliberately faulty infrastructure, especially when that infrastructure is absolutely necessary to keep out the devastation on the surface.

Here’s the pitch: A smaller and more intimate take on Fallout, styled after indie titles like Papers, Please. Each day would bring new items to repair and new pressures to consider. For instance, the vault’s quartermaster needs his air circulator fixed. You have the skills to do this, but your family also needs more nutrients than daily rations allow. Do you repair the unit and hope that your good faith sways him, or refuse to give his unit back until he delivers the needed food?

Repairs would gradually get more complex, and the needs in the vault would grow more dire. As more systems in the shelter failed, there’d be no way to satisfy everyone. Vault 53 demands you make sacrifices; where those sacrifices come from is up to you.

Vault 69

Vault 69 begs the question: was anyone at Vault-Tec ever a scientist?

The vault had the standard thousand residents, but only one of those residents was a man. Less the product of principled hypotheses and more that of a giggling middle schooler, Vault 69 has virtually no available information.

Here’s the pitch: Uhh, no comment on this one

This, but not a dog. A puppet.

Vault 77

Although we don’t have the name of the sole resident of Vault 77, we do have a significant amount of information on his circumstances. Locked into the shelter alone, it took more than a year for him to open the crate marked “Government Issue: Puppet Ration.” Once he did, he swiftly descended into a delusional and codependent relationship with the inanimate objects.

After developing an intimate bond with one in particular (a puppet sporting a Vault-Tec jumpsuit similar to his own), the man eventually left the vault. Accompanied by the puppet and a tamed giant ant named “Mr. Pinch,” his expedition into the wasteland was an unwilling miasma of violence and death, led by the whims of his puppet.

Here’s the pitch: Styled after A Way Out, this linear adventure would be a mandatory co-op experience. I’ll get this out of the way now; yes, the player assigned to the puppet would have to spend the first hour or so in a box. It’s important to the characters’ motivations. Once the two players are united however, it would largely be the puppet’s show.

The puppet could tell the human character where to walk, which quests to take, who to kill. Although the human may have locomotion, the puppet ironically holds all the power.

Fallout is a series that embraces both the devastation and the absurdity of post-apocalyptic life. It’d be great to see Bethesda to keep experimenting with the style of games in its unique universe. Just, for the love of god, stay away from Vault 69.


Head here to read our review of the first season of the Fallout TV show. And here are the lessons someone should (and shouldn’t) take from Fallout. For a more serious take, check out our exhaustive history on the making of Fallout 4.

Francis Fan Lee, former professor and interdisciplinary speech processing inventor, dies at 96

Francis Fan Lee, former professor and interdisciplinary speech processing inventor, dies at 96

Francis Fan Lee ’50, SM ’51, PhD ’66, a former professor of MIT’s Department of Electrical Engineering and Computer Science, died on Jan. 12, some two weeks shy of his 97th birthday.

Born in 1927 in Nanjing, China, to professors Li Rumian and Zhou Huizhan, Lee learned English from his father, a faculty member in the Department of English at the University of Wuhan. Lee’s mastery of the language led to an interpreter position at the U.S. Office of Strategic Services, and eventually a passport and permission from the Chinese government to study in the United States. 

Lee left China via steamship in 1948 to pursue his undergraduate education at MIT. He earned his bachelor’s and master’s degrees in electrical engineering in 1950 and 1951, respectively, before going into industry. Around this time, he became reacquainted with a friend he’d known in China, who had since emigrated; he married Teresa Jen Lee, and the two welcomed children Franklin, Elizabeth, Gloria, and Roberta over the next decade. 

During his 10-year industrial career, Lee distinguished himself in roles at Ultrasonic (where he worked on instrument type servomechanisms, circuit design, and a missile simulator), RCA Camden (where he worked on an experimental time-shared digital processor for department store point-of-sale interactions), and UNIVAC Corp. (where he held a variety of roles, culminating in a stint in Philadelphia, planning next-generation computing systems.)

Lee returned to MIT to earn his PhD in 1966, after which he joined the then-Department of Electrical Engineering as an associate professor with tenure, affiliated with the Research Laboratory of Electronics (RLE). There, he pursued the subject of his doctoral research: the development of a machine that would read printed text out loud — a tremendously ambitious and complex goal for the time.

Work on the “RLE reading machine,” as it was called, was inherently interdisciplinary, and Lee drew upon the influences of multiple contemporaries, including linguists Morris Halle and Noam Chomsky, and engineer Kenneth Stevens, whose quantal theory of speech production and recognition broke down human speech into discrete, and limited, combinations of sound. One of Lee’s greatest contributions to the machine, which he co-built with Donald Troxel, was a clever and efficient storage system that used root words, prefixes, and suffixes to make the real-time synthesis of half-a-million English words possible, while only requiring about 32,000 words’ worth of storage. The solution was emblematic of Lee’s creative approach to solving complex research problems, an approach which earned him respect and admiration from his colleagues and contemporaries.

In reflection of Lee’s remarkable accomplishments in both industry and building the reading machine, he was promoted to full professor in 1969, just three years after he earned his PhD. Many awards and other recognition followed, including the IEEE Fellowship in 1971 and the Audio Engineering Society Best Paper Award in 1972. Additionally, Lee occupied several important roles within the department, including over a decade spent as the undergraduate advisor. He consistently supported and advocated for more funding to go to ongoing professional education for faculty members, especially those who were no longer junior faculty, identifying ongoing development as an important, but often-overlooked, priority.

Lee’s research work continued to straddle both novel inquiry and practical, commercial application — in 1969, together with Charles Bagnaschi, he founded American Data Sciences, later changing the company’s name to Lexicon Inc. The company specialized in producing devices that expanded on Lee’s work in digital signal compression and expansion: for example, the first commercially available speech compressor and pitch shifter, which was marketed as an educational tool for blind students and those with speech processing disorders. The device, called Varispeech, allowed students to speed up written material without losing pitch — much as modern audiobook listeners speed up their chapters to absorb books at their preferred rate. Later innovations of Lee’s included the Time Compressor Model 1200, which added a film and video component to the speeding-up process, allowing television producers to subtly speed up a movie, sitcom, or advertisement to precisely fill a limited time slot without having to resort to making cuts. For this work, he received an Emmy Award for technical contributions to editing.

In the mid-to-late 1980s, Lee’s influential academic career was brought to a close by a series of deeply personal tragedies, including the 1984 murder of his daughter Roberta, and the subsequent and sudden deaths of his wife, Theresa, and his son, Franklin. Reeling from his losses, Lee ultimately decided to take an early retirement, dedicating his energy to healing. For the next two decades, he would explore the world extensively, a nomadic second chapter that included multiple road trips across the United States in a Volkswagen camper van. He eventually settled in California, where he met his last wife, Ellen, and where his lively intellectual life persisted despite diagnoses of deafness and dementia; as his family recalled, he enjoyed playing games of Scrabble until his final weeks. 

He is survived by his wife Ellen Li; his daughters Elizabeth Lee (David Goya) and Gloria Lee (Matthew Lynaugh); his grandsons Alex, Benjamin, Mason, and Sam; his sister Li Zhong (Lei Tongshen); and family friend Angelique Agbigay. His family have asked that gifts honoring Francis Fan Lee’s life be directed to the Hertz Foundation

Fostering research, careers, and community in materials science

Fostering research, careers, and community in materials science

Gabrielle Wood, a junior at Howard University majoring in chemical engineering, is on a mission to improve the sustainability and life cycles of natural resources and materials. Her work in the Materials Initiative for Comprehensive Research Opportunity (MICRO) program has given her hands-on experience with many different aspects of research, including MATLAB programming, experimental design, data analysis, figure-making, and scientific writing.

Wood is also one of 10 undergraduates from 10 universities around the United States to participate in the first MICRO Summit earlier this year. The internship program, developed by the MIT Department of Materials Science and Engineering (DMSE), first launched in fall 2021. Now in its third year, the program continues to grow, providing even more opportunities for non-MIT undergraduate students — including the MICRO Summit and the program’s expansion to include Northwestern University.

“I think one of the most valuable aspects of the MICRO program is the ability to do research long term with an experienced professor in materials science and engineering,” says Wood. “My school has limited opportunities for undergraduate research in sustainable polymers, so the MICRO program allowed me to gain valuable experience in this field, which I would not otherwise have.”

Like Wood, Griheydi Garcia, a senior chemistry major at Manhattan College, values the exposure to materials science, especially since she is not able to learn as much about it at her home institution.

“I learned a lot about crystallography and defects in materials through the MICRO curriculum, especially through videos,” says Garcia. “The research itself is very valuable, as well, because we get to apply what we’ve learned through the videos in the research we do remotely.”

Expanding research opportunities

From the beginning, the MICRO program was designed as a fully remote, rigorous education and mentoring program targeted toward students from underserved backgrounds interested in pursuing graduate school in materials science or related fields. Interns are matched with faculty to work on their specific research interests.

Jessica Sandland ’99, PhD ’05, principal lecturer in DMSE and co-founder of MICRO, says that research projects for the interns are designed to be work that they can do remotely, such as developing a machine-learning algorithm or a data analysis approach.

“It’s important to note that it’s not just about what the program and faculty are bringing to the student interns,” says Sandland, a member of the MIT Digital Learning Lab, a joint program between MIT Open Learning and the Institute’s academic departments. “The students are doing real research and work, and creating things of real value. It’s very much an exchange.”

Cécile Chazot PhD ’22, now an assistant professor of materials science and engineering at Northwestern University, had helped to establish MICRO at MIT from the very beginning. Once at Northwestern, she quickly realized that expanding MICRO to Northwestern would offer even more research opportunities to interns than by relying on MIT alone — leveraging the university’s strong materials science and engineering department, as well as offering resources for biomaterials research through Northwestern’s medical school. The program received funding from 3M and officially launched at Northwestern in fall 2023. Approximately half of the MICRO interns are now in the program with MIT and half are with Northwestern. Wood and Garcia both participate in the program via Northwestern.

“By expanding to another school, we’ve been able to have interns work with a much broader range of research projects,” says Chazot. “It has become easier for us to place students with faculty and research that match their interests.”

Building community

The MICRO program received a Higher Education Innovation grant from the Abdul Latif Jameel World Education Lab, part of MIT Open Learning, to develop an in-person summit. In January 2024, interns visited MIT for three days of presentations, workshops, and campus tours — including a tour of the MIT.nano building — as well as various community-building activities.

“A big part of MICRO is the community,” says Chazot. “A highlight of the summit was just seeing the students come together.”

The summit also included panel discussions that allowed interns to gain insights and advice from graduate students and professionals. The graduate panel discussion included MIT graduate students Sam Figueroa (mechanical engineering), Isabella Caruso (DMSE), and Eliana Feygin (DMSE). The career panel was led by Chazot and included Jatin Patil PhD ’23, head of product at SiTration; Maureen Reitman ’90, ScD ’93, group vice president and principal engineer at Exponent; Lucas Caretta PhD ’19, assistant professor of engineering at Brown University; Raquel D’Oyen ’90, who holds a PhD from Northwestern University and is a senior engineer at Raytheon; and Ashley Kaiser MS ’19, PhD ’21, senior process engineer at 6K.

Students also had an opportunity to share their work with each other through research presentations. Their presentations covered a wide range of topics, including: developing a computer program to calculate solubility parameters for polymers used in textile manufacturing; performing a life-cycle analysis of a photonic chip and evaluating its environmental impact in comparison to a standard silicon microchip; and applying machine learning algorithms to scanning transmission electron microscopy images of CrSBr, a two-dimensional magnetic material. 

“The summit was wonderful and the best academic experience I have had as a first-year college student,” says MICRO intern Gabriella La Cour, who is pursuing a major in chemistry and dual degree biomedical engineering at Spelman College and participates in MICRO through MIT. “I got to meet so many students who were all in grades above me … and I learned a little about how to navigate college as an upperclassman.” 

“I actually have an extremely close friendship with one of the students, and we keep in touch regularly,” adds La Cour. “Professor Chazot gave valuable advice about applications and recommendation letters that will be useful when I apply to REUs [Research Experiences for Undergraduates] and graduate schools.”

Looking to the future, MICRO organizers hope to continue to grow the program’s reach.

“We would love to see other schools taking on this model,” says Sandland. “There are a lot of opportunities out there. The more departments, research groups, and mentors that get involved with this program, the more impact it can have.”