Hacking Neural Nets

FromLinear Digressions

Start listening View podcast show

Hacking Neural Nets

FromLinear Digressions

ratings:

Length:

15 minutes

Released:

Jan 5, 2016

Format:

Podcast episode

Description

Machine learning: it can be fooled, just like you or me. Here's one of our favorite examples, a study into hacking neural networks.

Relevant links:
http://arxiv.org/pdf/1412.1897v4.pdf

Released:

Jan 5, 2016

Format:

Podcast episode

Titles in the series (100)

Linear Digressions is a podcast about machine learning and data science. Machine learning is being used to solve a ton of interesting problems, and to accomplish goals that were out of reach even a few short years ago.

Skip carousel

Um Detector 1: So, um... what about machine learning for audio a… by Linear Digressions
13 min listen
Hunting for the Higgs: Machine learning and particle physics go together… by Linear Digressions
10 min listen
Stats of World Series Streaks: Baseball is characterized by a high level of equa… by Linear Digressions
13 min listen
Statistical Mistakes and the Challenger Disaster: After the Challenger exploded in 1986, killing al… by Linear Digressions
13 min listen
Labels and Where To Find Them: Supervised classification is built on the backs o… by Linear Digressions
13 min listen
Random Kanye: Ever feel like you could randomly assemble words … by Linear Digressions
9 min listen
The Enron Dataset: In 2000, Enron was one of the largest and compani… by Linear Digressions
12 min listen
Genetics and Um Detection (HMM Part 2): In part two of our series on Hidden Markov Models… by Linear Digressions
15 min listen
Electoral Insights (Part 2): Following up on our last episode about how experi… by Linear Digressions
21 min listen
Computers Try to Tell Jokes: Computers are capable of many impressive feats, b… by Linear Digressions
9 min listen
Lie Detectors: Often machine learning discussions center around … by Linear Digressions
9 min listen
Electoral Insights (Part 1): The first of our two-parter discussing the recent… by Linear Digressions
9 min listen
Introducing Hidden Markov Models (HMM Part 1): Wikipedia says, "A hidden Markov model (HMM) is a… by Linear Digressions
15 min listen
How Outliers Helped Defeat Cholera: In the 1850s, there were a lot of things we didn’… by Linear Digressions
11 min listen
Backpropagation: The reason that neural nets are taking over the w… by Linear Digressions
12 min listen
Neural Nets (Part 2): In the last episode, we zipped through neural net… by Linear Digressions
11 min listen
Reporter Bot: There’s a big difference between a table of numbe… by Linear Digressions
11 min listen
Facial Recognition with Eigenfaces: A true classic topic in ML: Facial recognition is… by Linear Digressions
10 min listen
Better Facial Recognition with Fisherfaces: Now that we know about eigenfaces (if you don't, … by Linear Digressions
12 min listen
Kalman Runners: The Kalman Filter is an algorithm for taking nois… by Linear Digressions
15 min listen
Watson: This machine learning algorithm beat the human ch… by Linear Digressions
16 min listen
Modeling Particles in Atomic Bombs: In a fun historical journey, Katie and Ben explor… by Linear Digressions
16 min listen
Falsifying Data: In the first of a few episodes on fraud in electi… by Linear Digressions
18 min listen
Thresholdout: Down with Overfitting: Overfitting to your training data can be avoided … by Linear Digressions
16 min listen
Random Number Generation: Let's talk about randomness! Although randomness … by Linear Digressions
10 min listen
Benford's Law: Sometimes numbers are... weird. Benford's Law is… by Linear Digressions
18 min listen
Bayesian Psychics: Come get a little "out there" with us this week, … by Linear Digressions
12 min listen
Inferring Authorship (Part 2): Now that we’re up to speed on the classic author … by Linear Digressions
14 min listen
Yiddish Translation: Imagine a language that is mostly spoken rather t… by Linear Digressions
12 min listen
Monte Carlo For Physicists: This is another physics-centered podcast, about a… by Linear Digressions
8 min listen

More Episodes from Linear Digressions

Skip carousel

So long, and thanks for all the fish: All good things must come to an end, including th… by Linear Digressions
36 min listen
A Reality Check on AI-Driven Medical Assistants: The data science and artificial intelligence comm… by Linear Digressions
14 min listen
A Data Science Take on Open Policing Data: A few weeks ago, we put out a call for data scien… by Linear Digressions
24 min listen
Procella: YouTube's super-system for analytics data storage: This is a re-release of an episode that originall… by Linear Digressions
30 min listen
The Data Science Open Source Ecosystem: Open source software is ubiquitous throughout dat… by Linear Digressions
23 min listen
Rock the ROC Curve: This is a re-release of an episode that first ran… by Linear Digressions
16 min listen
Criminology and Data Science: This episode features Zach Drake, a working data … by Linear Digressions
31 min listen
Racism, the criminal justice system, and data science: As protests sweep across the United States in the… by Linear Digressions
32 min listen
An interstitial word from Ben: A message from Ben around algorithmic bias, and h… by Linear Digressions
6 min listen
Convolutional Neural Networks: This is a re-release of an episode that originall… by Linear Digressions
22 min listen
Stein's Paradox: This is a re-release of an episode that was origi… by Linear Digressions
27 min listen
Protecting Individual-Level Census Data with Differential Privacy: The power of finely-grained, individual-level dat… by Linear Digressions
21 min listen
Causal Trees: What do you get when you combine the causal infer… by Linear Digressions
15 min listen
The Grammar Of Graphics: You may not realize it consciously, but beautiful… by Linear Digressions
36 min listen
Gaussian Processes: It’s pretty common to fit a function to a dataset… by Linear Digressions
21 min listen
Keeping ourselves honest when we work with observational healthcare data: The abundance of data in healthcare, and the valu… by Linear Digressions
19 min listen
Changing our formulation of AI to avoid runaway risks: Interview with Prof. Stuart Russell: AI is evolving incredibly quickly, and thinking n… by Linear Digressions
29 min listen
Putting machine learning into a database: Most data scientists bounce back and forth regula… by Linear Digressions
24 min listen
The work-from-home episode: Many of us have the privilege of working from hom… by Linear Digressions
29 min listen
Understanding Covid-19 transmission: what the data suggests about how the disease spreads: Covid-19 is turning the world upside down right n… by Linear Digressions
25 min listen
Network effects re-release: when the power of a public health measure lies in widespread adoption: This week’s episode is a re-release of a recent e… by Linear Digressions
27 min listen
Causal inference when you can't experiment: difference-in-differences and synthetic controls: When you need to untangle cause and effect, but y… by Linear Digressions
21 min listen
Better know a distribution: the Poisson distribution: This is a re-release of an episode that originall… by Linear Digressions
32 min listen
The Lottery Ticket Hypothesis: Recent research into neural networks reveals that… by Linear Digressions
20 min listen
Interesting technical issues prompted by GDPR and data privacy concerns: Data privacy is a huge issue right now, after yea… by Linear Digressions
20 min listen
Thinking of data science initiatives as innovation initiatives: Put yourself in the shoes of an executive at a bi… by Linear Digressions
17 min listen
Building a curriculum for educating data scientists: Interview with Prof. Xiao-Li Meng: As demand for data scientists grows, and it remai… by Linear Digressions
32 min listen
Running experiments when there are network effects: Traditional A/B tests assume that whether or not … by Linear Digressions
25 min listen
Zeroing in on what makes adversarial examples possible: Adversarial examples are really, really weird: pi… by Linear Digressions
23 min listen
Unsupervised Dimensionality Reduction: UMAP vs t-SNE: Dimensionality reduction redux: this episode cove… by Linear Digressions
30 min listen

Discover this podcast and so much more

Hacking Neural Nets

Hacking Neural Nets

Description

Titles in the series (100)

More Episodes from Linear Digressions