Michael Jacobson

✅Have you ever thought of an alternative cost function for optimizing prosthesis assistance parameters? Well now you can read our most recent publication in @SciReports how we approached this challenge! @prakyath_k @RehabRoboLab 🧵 For more details nature.com/articles/s4159…

@pranavab47 @ErnestoHzHj @UMRobotics @doctorBobG Congratulations @ErnestoHzHj !

Congratulations to Ernesto Hernandez @ErnestoHzHj for successfully defending his PhD thesis. He worked on data-driven low-level control of bipeds and implemented his controllers on Digit. Wishing him all the best as he heads to @UMRobotics to work with @doctorBobG

@fcrameri Wow I wish I knew about this sooner! I will definitely be implementing & citing for future research.

Scientific colour maps 8.0 fabiocrameri.ch/colourmaps #useBatlow #ScientificVisualisation #visualisation #Science #ColourPalette #DataVis #GraphicDesign #OpenSource #Accessibility #Colour #AcademicCommunityContribution #ACC 1/n

@akshay_pachaar Is there an online book like this with a similar format to the interactive book Dive into Deep Learning?

Received "The Python Workshop" a few days ago! The book focuses on solving real-world problems using Python, enabling you to learn by doing! Topics cover: - Python Fundamentals - Software Development - Data Analytics - Machine Learning Highly recommended!! Check this out 👇

@prakyath_k @KarlZelik I’m glad this is being brought to light. There’s a lot to be learned from the strength & conditioning domain. Another reason to diversify perspectives and reduce tunnel-vision within your respective field.

@KarlZelik @Michael_J_0 right up your alley

5 years ago, I knew nothing about EQUIPPED powerlifting. But I really should have. If you're interested in occupational #exoskeletons or #biomechanics then you'll probably find this topic fascinating as well. It's pretty mind-blowing stuff! 😳 💪🏽 Here's what it is & why...

@KarlZelik Invest time and effort into projects outside of the standard course curriculum. My work while I was in the Society of Automotive Engineers helped set me apart from other applicants. I was able to convey my passion and drive to industry recruiters through these projects.

What is your best piece of advice for undergraduate students interested in finding industry internships? (ideally something actionable & specific)

@Artifexx @scite @LitMaps @zotero @obsdmd @prakyath_k

Here is my method to conduct (and automate) a literature review. Using Google Scholar, @scite , @litmaps , #ChatGPT, @zotero and @obsdmd. Tutorial with examples and best practices: 👇

@DaveShermanPT @13machine13 Have you also discussed once you have all of the data bucketed into every stride, then how to classify the data within the representative “average stride” ? All of the data will have different lengths, so you’re either choosing to interpolate or truncate the data.

@13machine13 Thanks for sharing! This debate started in my lab this week - most for segmenting strides or reps from IMU signals. A wide range of approaches all using some amplitude measure sensitive to signal to noise ratio. Easy to find in rate of torque development (above) when SNR is high

Friday Debate: How does your lab find this point?

@Andrescv18 @DaveShermanPT Agreed, you could use a higher order of the data and set a second condition which is a window or threshold.

@LabLiao @SICB_DCB_DVM Try @Protolabs they also have a quick turn around time.

Does anyone know a company that will 3-D print (or machine) molds from acrylic (or another UV transparent material)? #3Dprint @SICB_DCB_DVM

Digit stepping and balance control using a custom low-level controller Full Video: youtu.be/MDujbyE3AyE Work by Ernesto Hernandez @agilityrobotics @MIE_UIC @UICEngineering #robotics

@ProfSilver @Lnolasco10 @DeannaHGates @JBiomech Interesting to see the impact leg length has on walking features. We tune the height of our robotic prosthetic during fitting and again during walking trials. I guess we should’ve taken a closer look at these changes.

@humotech @walshharvard @SpringerNature @Harvard Completely agreed! Off-board actuation helps us rapidly test control algorithms with ease. I can also see a place for them in clinical settings.

Thanks @walshharvard for "Opportunities and challenges in the development of exoskeletons for locomotor assistance" in @SpringerNature. We wholeheartedly agree on the value of offboard actuation and love supporting this growing field with #Caplex! @Harvard zcu.io/3Nee

We've just released the first version of our Deep Learning Tuning Playbook! This is our attempt to distill our process for actually getting good results with deep learning. We emphasize hyperparameter tuning since it has been a large pain point. github.com/google-researc…

Year in Review: @MIE_UIC PhD candidate Michael Jacobson spent 2 months this summer working in Ukraine with Breaking the Chains International as they removed animals from dangerous situations and took them to safety. Read more about this remarkable story: shorturl.at/mos49

@hubermanlab @lexfridman As a researcher/military veteran who recently spent time living in dangerous parts of Ukraine, it was resonating to hear Lex’s opinions. Within the midst of widespread death/destruction, you start to unravel the genuine hate and lack of empathy bred from war itself. @lexfridman

.@lexfridman on the Huberman Lab podcast out now #Love hubermanlab.com/dr-lex-fridman…

#MuJoCo python tutorials aimed at beginners. These include simple examples that can be combined to build more complex simulations. tiny.cc/mujocopy

@ethanCaballero @prakyath_k

Which AI tech companies don't have hiring freeze for PhD student interns/student_researchers right now?

@p_franks_ @inseungkang @KarlZelik @doctorBobG @g_m_bryan Wow, I wonder who has the most time spent with their exo. @prakyath_k you're probably up there when it comes to squatting with an exo 😂

@inseungkang @Michael_J_0 @KarlZelik @doctorBobG @g_m_bryan 400 hours well spent 😅

If you're evaluating #exoskeletons (or reading studies that do), this may be the single most important thing to do 👇🏽 Before jumping to conclusions or overgeneralizing, ask yourself: 🤔 How much time were users given to acclimate to the exo (i.e., the new tool)? A short 🧵...

@inseungkang @KarlZelik @doctorBobG @p_franks_ @g_m_bryan Hmm, this is an interesting point. There are previous adaptation studies that have resulted in quicker 'learning'. Again, in the case that we do not have an unlimited experimental time budget, considering 'short-term learning' may be precedent for rehabilitation goals.

@Michael_J_0 @KarlZelik @doctorBobG Yes, 'learning' how to effectively use an exo is important. I remember having a chat with @p_franks_ @g_m_bryan asking about the number of hours they've walked in their exos during grad school. But, it is worth noting that adaptation does not always lead to learning.