al0x23

dream = securityResearcher && mathematician; // but life ain't a bool

[al0x23 @al0x23]

It was a long journey till the first meaningful result… would have taken even longer without @PatrickAlphaC @Jeyffre @0xriptide free resources and daily dose of motivation. A big thanks also to @GalloDaSballo who kindly answered all my questions when I was starting 3 months ago

[al0x23 @al0x23]

I've had some mental health issues in the last 2 months, feeling very unmotivated and sad on a daily basis, couldn't even sleep at night. Need some rest, I hope to come back soon, in the meantime I hope you can all have success in web3!

[al0x23 @al0x23]

day 0x02d of my 0x423 challenge

Daily reflection -> rounding errors and real-world consequences

In Web3 security, we know that rounding errors can mean unbounded losses for a protocol, but today I was reminded how much more is at stake in other contexts too. On February 25, 1991, during the Gulf War, a U.S. Patriot missile battery in Dhahran, Saudi Arabia, failed to intercept an incoming Iraqi Scud missile due to a rounding error in computer arithmetic. The Scud missile struck an American Army barracks, killing 28 soldiers and injuring around 100 others. The root cause was a software flaw in the system’s time calculation. The system used a 24-bit fixed-point register to store time in tenths of a second. To convert this to seconds, it multiplied the value by 1/10. But 1/10 cannot be represented exactly in binary, so each conversion introduced a tiny rounding error. The system had been running for about 100 hours, and the small error accumulated to about 0.34 seconds.A Scud travels at roughly 1,676 m/s, so in 0.34 seconds it moves over half a kilometer — enough to leave the system’s tracking “range gate,” causing the Patriot to miss. It’s a powerful reminder that tiny floating-point errors, when accumulated over time in real-time systems, can lead to catastrophic failures.

• MST124 — Essential Mathematics 1
Finished Section 1 on types of numbers
Started Section 2 of Module 1 on algebraic expressions

• M140 — Introducing Statistics
Finished Section 3 of Unit 1 on preparing data for analysis

• Rust — RareCode
Completed Journey 58 on enum attributes!

[al0x23 @al0x23]

Going carefully through missed issues is the path to success for SRs… but it is as useful as it’s painful

[al0x23 @al0x23]

day 0x02c of my 0x423 challenge (days on/off ratio = 6.29)

Sooo, the big news is that, as programmed, I’m starting my math degree with @OpenUniversity (fully remote and part time) and I’m very thankful for having this opportunity to study mathematics more deeply. My first modules are MST124 Essential Mathematics 1 and M140 Introducing Statistics.
Since I’m also auditing many hours per day as my main task and I'd like to avoid burnout, I’m putting the ZK Book path on pause, while I’ll try to finish Maths for Machine Learning on MathAcademy and RareCode at a slower pace. The Computer Science / AI track should remain… We’ll see how the next contests go from this moment on!

Fun fact: by the time this challenge ends, I should already hold a bachelor in mathematics… which would be more than awesome!!

• MST124 — Essential Mathematics 1
Started reading the first chapter. Although it’s pretty basic mathematics, there is always something new to learn if you look carefully… a bit like with bugs!

• M140 — Introducing Statistics
Started the first chapter here too, on how to clean data, e.g. spurious precision, when the additional digits of what you are measuring give a false sense of precision + general intro to statistical analysis.

[al0x23 @al0x23]

day 0x02b of my 0x423 challenge (days on/off ratio 6.14)

Tomorrow there are going to be major changes in the study plan... can't wait to write tomorrow's learning post!!

• Math Academy — 64/60 XP (59% Mathematics for Machine Learning)

open and closed sets (review)
writing sums of trigonometric functions in amplitude–phase form (review)
the intersection of two planes (review)
the constant multiple rule for indefinite integrals
eccentricity and vertices of hyperbolas (review)
constrained optimization of quadratic forms: determining where extrema are attained (review)
orthogonal matrices (review)
the antiderivative
elliptic cones (review)
solving systems of equations using inverse matrices (review)
the hyperbolic functions (review)

[al0x23 @al0x23]

@iconthegreatt why not?? 😌

[al0x23 @al0x23]

Seems a very nice contest (+ with the best judge around…)

Who is going to join me in the unlocking mission?? 👊

[JanuaryPersimmon2024 @AifosSi]

Super happy and grateful for this upcoming judging opportunity!!

Unlock the Crit/High pot and grab 1st place anon! 🫵🫵

[al0x23 @al0x23]

day 0x02a of my 0x423 challenge (days on/off ratio: 6)

Having troubles in not making mistakes while doing simple multiplications/additions for singular value decomposition of a larger matrix with maybe not orthogonal vectors and consequent Gram-Schmidt process... 1 XP = 1 minute of focused work is definitely not the case for me here

• Math Academy — 57/60 XP (59% Mathematics for Machine Learning)

singular value decomposition of larger matrices
identifying quadric surfaces (review)
identifying a rational function from a graph (review)
tree diagrams for independent events (review)

[al0x23 @al0x23]

@iconthegreatt A lot ahah I think it’s the best way to make quick progress with 1 hour per day or even less, and get exposure to many different topics (and many exercises, although a bit repetitive sometimes) !

[al0x23 @al0x23]

day 0x029 of my 0x423 challenge (days on/off ratio 5.86)

I really do start enjoying diagonalizing matrices, which is basically all I've done in this course till now ahah

• Math Academy — 68/60 XP (59% Mathematics for Machine Learning)

singular value decomposition of 2×2 matrices
singular value decomposition of 2×2 matrices with zero or repeated eigenvalues
diagonalization of 3×3 symmetric matrices (review)
the Cauchy–Schwarz inequality and the angle between two vectors (review)
the Fundamental Theorem of Algebra with quartic equations (review)

[al0x23 @al0x23]

day 0x028 of my 0x423 challenge ( days on/off ratio 5.71)

• Math Academy — 40/60 XP (57% Mathematics for Machine Learning)

• ZK Book
Played with Circom and created my first circuit and witness locally.

• CS / AI Fundamentals — MIT 6.042
Continued Problem Set 4, solved some interesting exercises on graph isomorphisms.

• Rust — RareCode
Completed Journey 57 on enum.

[al0x23 @al0x23]

day 0x027 of my 0x423 challenge (on/off days ratio = 5.57)

• Math Academy — 65/60 XP (57% Mathematics for Machine Learning)
the empirical rule for the normal distribution (review)
the Remainder Theorem (review)
modeling financial problems using geometric series (review)
the singular values of a matrix
variance of sample means (review)
the least-squares solution of a linear system (without collinearity) (review)
the Gram–Schmidt process for two vectors (review)

• ZK Book
Read the “Hello Circom” chapter.

• CS / AI Fundamentals — MIT 6.042
None.

• Rust — RareCode
Completed Journey 56, very useful match statements.

[al0x23 @al0x23]

@KrisRenzo If I were you, I’d try more proud of myself for finding them… that’s the most important skill. And luck might come back with interests later!

[al0x23 @al0x23]

@CarterToB Didn’t know you had the extra usage option, with slippage protection too! 🤩

[al0x23 @al0x23]

day 0x026 of my 0x423 challenge (days on/off ratio: 7.6)

Today was a bit taken by an audit, but still managed to progress in my math journey!

• Math Academy — 45/60 XP (57% Mathematics for Machine Learning)
constrained optimization of quadratic forms: determining where extrema are attained
linear regression with matrices (review)
parametric equations of horizontal hyperbolas (review)
projecting vectors onto subspaces in Euclidean spaces (arbitrary bases): applications (review)
introduction to principal component analysis

[al0x23 @al0x23]

@heezeEth @RareCodeAI Congrats for making it!!

[al0x23 @al0x23]

day 0x025 of my 0x423 challenge (on/off days ratio = 9.25)

• Math Academy — 66/60 XP (56% Mathematics for Machine Learning)

the sample covariance matrix
the least-squares solution of a linear system (with collinearity) (review)
log-likelihood functions for discrete probability distributions
the sample variance
diagonalization of 3×3 symmetric matrices

• ZK Book
Went through the Tornado Cash implementation and read the introduction to ZK circuits with Circom.

• CS / AI Fundamentals — MIT 6.042
None.

• Rust — RareCode
Completed Journey 55 on the Result type.

[al0x23 @al0x23]

@0xFlint_ @pashov mod 18

[al0x23 @al0x23]

day 0x024 of my 0x423 challenge (days off = 4, days on/off ratio = 9)

• Math Academy — 68/60 XP (55% Mathematics for Machine Learning)

diagonalization of 2×2 symmetric matrices
identifying quadric surfaces
constrained optimization of quadratic forms
cylinders
orthogonal linear transformations

• ZK Book
Finished the review; tomorrow → Tornado Cash implementation.

• CS / AI Fundamentals — MIT 6.042
Finished Lecture 7 and started Problem Set 4.

• Rust — RareCode
Completed Journey 54: never going to try to print out a grid as a string in my life after this one…