Earlier last week I posted a poll on twitter asking If my readers would like me to post a GPT generated article. The votes were very evenly distributed:

The remainder of this article is generated using GPT-2 network (using this site) primed on bits of my other articles to covey some of the style. The images were generated by https://app.generative.photos/ from RosebudAI - a recent hot startup in the AI space. When done reading, please consider future historians analyzing the outburst of AI in 2010-2020 and decide if they'd be impressed or will they be like "WTF were they thinking back then!?". 

The study was done in the summer of 2014, but there have been so many recent news stories about Uber (and similar companies) and the impact it has had on public safety, ”We're very happy” to add to the body of knowledge we've accumulated.

What can we learn about the state of public transportation?

Our findings indicate that if public transportation is to be made safe, “we have to build the systems on a much higher level”, and that this will require substantial change from the traditional public-sector perspective. We've discussed the problems in the above graphic:

In … Read more...

Welcome back. First of all, apologies for not posting as frequently as I used to. As you might imagine, blogging is not my full time job and I'm currently extremely involved in a very exciting startup (something I'm going to write about soon). On weekends and evening I'm busy with 7mo infant to help care for and altogether that leaves me with very little time. But I'll try to make it better soon, since a lot is going on in the AI space and signs of cooling are visible now all over the place.

In this post I'd like to focus on the recent book by Gary Marcus and Ernest Davis, Rebooting AI. Let's jump in.

If you are a person who is not necessarily deeply involved in recent (recent 10 years or so) developments in AI and instead you've been building your image of the field based on flashy PR statements by various big companies (including Google, Facebook, Intel, IBM and numerous smaller players) - this is a book for you. The first part of the book goes thoroughly through various press releases and "revolutionary" products and tracks how these projects either spectacularly or quietly failed. 

Reading the first … Read more...

Introduction

It's been roughly a year since I posted my viral "AI winter is well on its way" post and like I promised I'll periodically post an update on the general AI landscape. I posted one some 6 months ago and now is time for another one. And there has been a lot of stuff going on lately and none of it has changed my mind - the AI bubble is bursting. And as with every bubble bursting we are in a blowoff phase in which those who have the most to lose are pulling out the most outrageous confidence pumping pieces they could think of, the ultimate strategy to con some more naive people to give them money. But let's go over what has been going on.

The serious stuff

Firstly let's go over the non-comical stuff. Three of the founding fathers of deep learning - Geoffrey Hinton, Yoshua Bengio and Yann Lecun - received a Turing award - the most prestigious award given out in computer science. If you think that I will somehow question this judgement you will be disappointed, I think deep learning is well worth the Turing award. The one thing that in … Read more...

Many people these days are fascinated by deep learning, as it enabled new capabilities in many areas, particularly in computer vision. Deep nets are however black boxes and most people have no idea how they work (and frankly most of us, scientists trained in the field can't tell exactly how they work either). But the success of deep learning and a set of its surprising failure modes teach us a valuable lesson about the data we process.

In this post I will present a perspective of what deep learning actually enables, how it relates to classical computer vision (which is far from being dead) and what are the potential dangers of relying on DL for critical applications.

The vision problem

First of all, some things need to be said about the problem of vision/computer vision. In principle it could be formulated as follows: given an image from a camera allow the computer to answer questions about the contents of that image. Such questions can range from "is there a triangle in the image", "is there a human face in the image" to more complex instances such as "is there a dog chasing a cat in the image". Although many of … Read more...

It has became a tradition that I write a quick update on the state of self driving car development every year when the California DMV releases their disengagement data [ 2017 post here, 2018 post here]. 2018 was an important year for self driving as we had seen the first fatal accident caused by an autonomous vehicle (the infamous Uber crash in Arizona).

Let me start with a disclaimer: I plot disengagements against human crashes and fatalities not because it is a good comparison, but because this is the only comparison we have. There are many reasons why this is not the best measure and depending on the reason the actual "safety" of AV may be either somewhat better or significantly worse than indicated here. Below are some of my reasons:

• A disengagement is a situation in which a machine cannot be trusted and the human operator takes over to avoid any danger. The precise definition under California law is:
  “a deactivation of the autonomous mode when a failure of the autonomous technology is detected or when the safe operation of the vehicle requires that the autonomous vehicle test driver disengage the autonomous mode and take immediate manual

Every rule of thumb in data science has a counterexample. Including this one. 

In this post I'd like to explore several simple and low dimensional examples that expose how our typical intuitions about the geometry of data may be fatally flawed. This is generally a practical post, focused on examples, but there is a subtle message I'd like to provide. In essence: be careful. It is easy to make data based conclusions which are totally wrong.

Dimensionality reduction is not always a good idea

It is a fairly common practice to reduce the input data dimension via some projection, typically via principal component analysis (PCA) to get a lower-dimensional, more "condensed" data. This often works fine, as often the directions along which data is separable align with the principal axis. But this does not have to be the case, see a synthetic example below:

from sklearn.neural_network import MLPClassifier
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA
from scipy.stats import ortho_group
from mpl_toolkits.mplot3d import Axes3D
import numpy as np
import matplotlib.pyplot as plt


N = 10 # Dimension of the data
M = 500 # Number of samples

# Random rotation matrix
R = ortho_group.rvs(dim=N)
# Data variances
variances = np.sort(np.random.rand((N)))[::-1]