Miron's Weblog

USENIX Security Conference 2011

miron — Fri, 12 Aug 2011 17:22:51 +0000

I am attending the USENIX security conference this week. Sessions are available online. Here are my notes from sessions that I found interesting (bold for extra):

Network Security in the Medium Term: 2061–2561 AD, Charles Stross

Stross is one of my favorite science fiction authors. The main direction of the talk was the future political importance of information security. This is due to the intrusiveness of future information breaches once lifelogging, bioinformatics and other very intimate technologies are adopted.

Fast and Precise Sanitizer Analysis with BEK, Pieter Hooimeijer, et al

Compared different HTML sanitizers using an automated harness. Sanitizers from MS included (4), as well as new implementations (3).
Four of these were equivalent.
Only one protected against all the examples from the XSS Cheat Sheet

Toward Secure Embedded Web Interfaces, Baptiste Gourdin, et al

50 security vulnerabilities reported to CERT
All manufacturers had vulnerabilities (XSS, CSRf, …)
Author proposes WebDroid security distribution for embedded web interfaces (framework as “firewall”)

Comprehensive Experimental Analyses of Automotive Attack Surfaces, Stephen Checkoway, et al

Cars have an instrument bus
access to the bus gives complete control
can disable breaks, engine, even while in motion
attack surface:
- bus extends to media ports and charging
- bluetooth
- remote keyless entry
- wifi
- digital radio
- telematics: automated crash reporting / roadside assistance
completely compromised by author: bluetooth, media ports, more
- crafted cdrom (iso-9660, wma)
- strcpy in bluetooth stack – craft trojan in android
- bruteforce pin for pairing (hours)
- undetectable to users
- telematics compromise through cellular interface
compromise is silent and attacks can be triggered later

Privacy in the Age of Augmented Reality, Alessandro Acquisti, et al

De-identified faces matched to identified (FB, …)
10% of FB profiles are pseudonymous
Experiments:
- Unidentified: dating site photos
- Identified: FB profiles
- Only match to highest ranked matched from matching algorithm:
- 10% success rate for re-identification
- Against pittpatt (acquired by google)
- 30% success
- Predicting SSN from DOB, etc.
- 5 digits matched in four attempts
- iPhone app for real-time re-identification
PPI – personally predictable information

Secure In-Band Wireless Pairing, Shyamnath Gollakota, et al

Authors present a method for secure wireless pairing
No secondary channel (display, keyboard, infrared, …)
secure against MITM
Tamper evident message
cannot be altered, hidden, prevented without being detected
patterns of silence based on hash of message
sync pattern longer than any collision

TRESOR Runs Encryption Securely Outside RAM, Tilo Müller and Felix C. Freiling:

Prevents cold book attack
Uses AES-NI instruction set

A Study of Android Application Security, William Enck, et al

Decompiled and statically analyzed 21 millions lines of free Android apps
Pervasive misuse of private info and bad security practices
High market penetration of ad networks

Permission Re-Delegation: Attacks and Defenses, Adrienne Porter Felt, et al

I call this “cross app request forgery”
A large fraction of apps mistakenly expose sensitive functionality through intents
Malicious apps can abuse this
For example, turn on BT, Wifi, GPS
Suggests a way to mitigate through dynamic privilege reduction

Telex: Anticensorship in the Network Infrastructure, Eric Wustrow, et al

Telex converts innocuous, unblocked websites into proxies, without their explicit collaboration
Trigger routing to a proxy while accessing an “innocent” web site by putting a special nonce in the TLS negotiation
Could be used to bypass state censorship
To be deployed by ISPs on routers
Idea: consider deploying on web servers

Three Researchers, Five Conjectures: An Empirical Analysis of TOM-Skype Censorship and Surveillance, Jeffrey Knockel, et al

Detailed analysis of Chinese censorship through the compromised version of Skype used in China (with Skype’s cooperation)
Application uses a list of keywords to flag conversations for surveillance
Keywords triggering surveillance include mostly political and location words
Conjectures by authors include:
“Censorship is effective, despite attempts to evade it.” ,
“Censored memes spread differently than uncensored memes.”,
“Keyword based censorship is more effective when the censored keywords are unknown and on-line activity is, or is believed to be, under constant surveillance.”,
“The types of keywords censored in peer-to-peer communications are fundamentally different than the types of keywords censored in client-server communications.”,
“Neologisms are an effective technique in evading keyword based censorship, but censors frequently learn of their existence.”
Complete lists with translations of the censorship and surveillance keywords for TOM-Skype are available at http://cs.unm.edu/~jeffk/tom-skype/

BuddyNS

miron — Tue, 25 Jan 2011 21:53:53 +0000

BuddyNS is a free secondary name service. Yes, free. I started using it a couple of months ago and had no issues.

Good for your random project domains where you can’t justify spending on DNS fees.

Max More is now CEO of Alcor

miron — Sun, 26 Dec 2010 06:59:29 +0000

It’s interesting to see one of the major characters in the Extropian movement (precursor to the H+ movement) become the CEO of Alcor.

Moving your Android Contact List to a New Phone

miron — Mon, 20 Dec 2010 04:31:25 +0000

This is a somewhat technical article and assumes knowledge of Android and Linux.

Just got a Nexus S, and had some issues moving my contact list from my old phone. So I decided to write this up.

You have two options:

* If you come from a ROM that allows export to SD, just use Import/Export to USB storage, copy the file over, then import it

* Option #2 would have been to use Titanium Backup. However, it doesn’t seem to work right for restoring on the Nexus S (yet).

* Otherwise, you can copy the contacts2.db file. Of course, you have to root your target phone first. Then copy the db file to the sdcard.

As root, do (assuming standard layout):

cd /data/data/com.android.providers.contacts/databases rm contacts2.db cat /sdcard/contacts2.db > contacts2.db chmod 660 contacts2.db ls -l .. # see who owns this directory chown contats2.db
You might have to restart your phone for the contacts to be re-read.

Atomically Precise Fabrication

miron — Fri, 12 Nov 2010 06:56:38 +0000

Zyvex can now build atomically precise 3-D structures from silicon. That’s a nano equivalent to the MakerBot.

Arbitrary structures can be used to build templates and tools that can further build other tools, bootstrapping a new industry.

Brain Emulation by 2030

miron — Sat, 21 Aug 2010 01:11:24 +0000

Over the past few years I’ve been thinking about whole brain emulation (WBE) and the required computational resources. My conclusion is that the required technology level will be reached in the 2025 – 2030 time frame.

Although most estimates focus on calculations per second, the relevant parameters are:

Calculations per second
Memory size
Memory bandwidth per node
Inter-node communication bandwidth

Here I assume the WBE detail level to be somewhere between level 4 (spiking neural model) and level 5 (electrophysiological model) from the Whole Brain Emulation Roadmap. The computational capacity required would be around 100 Exa-FLOPS (1e20) and the memory capacity 10 Peta-bytes (1e13). Plugging these into the CPU and memory timeline calculators, gives an expected arrival time of 2026 and 2020 respectively for $1M.

Based on my models, it appears that memory bandwidth will be a significant bottleneck, while inter-CPU node bandwidth will not be.

I’ve created a spreadsheet where the two bandwidths are estimated.

The architecture I envision is 1 million compute nodes arranged in a 2-D cluster, 1000 on the side. Each node will have 100 TFLOPS CPU capacity and 10 GB of memory. Nodes are connected to the nearest 4 neighbors using a high speed bus. Longer distances are covered by multiple hops.

The inter-node bandwidth is dependent on the distance that voltage information has to travel, which is related to the axon length. Although some axons are very long, axon lengths probably follow a power law distribution, and long ones are rare. Based on a 1KHz update frequency and ~60 nodes within average axon distance, the required communication bandwidth is 250 Gb/s, which is close to the capabilities of current technology.

The local node memory bandwidth is dependent on the amount of synapse state memory and required refresh rate. Based on a 1KHz update rate, this works out to 14 TB/s. This figure is about 3 orders of magnitude higher than current technology.

The high memory bandwidth required makes it is likely that some kind of CPU/memory on-die integration will be required. The memristor seems a good candidate.

It is interesting to think of the brain as a 2.x dimension structure, due to the non-zero depth and also due to the folds. The question has been asked if this provides a barrier to implementation. The depth aspect is not an issue, since dividing the brain on a 2-D grid will put most vertical columns on the same node. Folds effectively reduce the length that an axon has to travel. It seems that in the worst case two points that “should” be a meter apart (brain diameter when laid flat) are only 10 cm apart (brain actual diameter). This may increase the inter-node bandwidth by a factor of 100, if most axons take advantage of the added dimensionality for routing. Since the inter-node bandwidth is relatively modest, it is doubtful that even a 100 fold increase will make it the bottleneck instead of the memory bandwidth.

In sum, here are the performance requirements:

1 million compute nodes in a 2-D topology
100 TFLOPS per node
10GB memory per node
10TB/s memory bandwidth
4 links to neighboring nodes at 250 Gb/s

It seems likely that this kind of machine will be available by 2030 for $1M, and possibly as early as 2025. It would be useful to evaluate the increase in memory bandwidth over time and extrapolate to this time frame.

Singularity Summit 2010 – live blogging – day 2

miron — Sun, 15 Aug 2010 17:59:11 +0000

Missed Eliezer Yudkowsky: Simplified Humanism and Positive Futurism.

9:40 – Ramez Naam: The Digital Biome

Plenty of carrying capacity for the biome – 30-300 billion people with advanced biotech. We are using only 1/1000 of the incident energy from the sun. There’s no reason to crash due to lack of resources with advanced tech. Population is predicted to level off at 10 billion.

Good points, but the Singularity is likely to happen on a shorter time scale.

10:40 – Lance Becker: Modifying the Boundary between Life and Death

Cells that experience ischemia (lack of oxygen) die after re-perfusion with oxygen, rather than during the ischemia itself. They die due to oxidative damage and apoptosis (cell suicide) related to mitochondria. How do we prevent the re-perfusion injury?

Current approaches – CPR, AED, Cooling. Next step cardiovascular bypass, mitochondria meds, better cooling, controlled reperfusion.

Tried cocktail of meds and cooling on 6 people that failed CPR, etc., and 50% were resuscitated.

Might be applications to Cryonics.

11:15 – Ellen Heber-Katz: The MRL mouse – how it regenerates and how we might do the same

Wound healing – scar formation (mammals) vs. epimorphic regeneration.

In regeneration – blastema – cell de-differentiation -> new limb. In mammals – regeneration in deer antlers. Rabbit ears.

MRL mouse – they regenerate various tissues. Inflammation response is necessary for regeneration. The mice don’t use mitochondria (?!). (during regeneration?) DCA blocks glycolysis – blocks regeneration.

MRL mouse cells are lacking p21 molecule. p21 protects against DNA damage, but blocks regeneration. Selective blocking of p21 is currently investigated. Removing p53 increases regeneration.

What about immortality? MRL mouse variant lives to 3 years, die of autoimmune response. p21 doesn’t get cancer until 16 months of age. But they get an inflammatory autoimmune disease – die of tumors.

Separate tumors, inflammation and regeneration.

11:50 – Anita Goel: Information Processing & Physical Intelligence in Nanomachines that Read/Write DNA

She is talking about controlling ribosomes and such from the macroscale.

Applications – includes “gene-radar” – portable diagnostics for DNA/RNA – point of care. Also, energy transduction at the nanoscale, read/write DNA, store information, computing.

DNA Polymerase can go forward, backward (error correction), or pause. This can be controlled from macro-scale. Quality of output can thus be controlled.

These machines are very energy efficient. How do they do that?

More stress on infrastructure-free diagnosis, developing world, etc. .

14:00 – Shane Legg: Universal measures of intelligence

Yet another exponential computation graph.

2025 – 10^20 ops per second. What does this translate to in meaningful intelligence? He wants intelligence on the Y axis.

Human vs. Ideal. Internal properties (“understanding”, “soul”, etc.) vs. External Properties. He is going to focus on External Ideal quadrant.

Definition of intelligence: “Intelligent system are expected to work, and work well, in many different environments.” – Gudwin

He shows an equation that takes into account success, agent, environment and Occam’s razor.

This measure is called Algorithmic Intelligence Quotient (AIQ). It seems to order some relevant learning agents in a correct way.

I think it would be great if we had a general intelligence measure, so we can gauge our success in creating artificial systems.

14:40 – John Tooby: Can discovering the design principles governing natural intelligence unleash breakthroughs in artificial intelligence?

For scientific debates, history shows us that the truth typically lies incredibly far beyond the limits of what even the most radical scientists of an era propose (e.g. Galileo and the size of the universe).

All the pieces are in place for realizing the Enlightenment project of a rigorous natural science of human nature: reverse engineering the code.

Emphasize evolution – natural selection is the only known physical process that can push against entropy. It only builds structures that solve evolutionarily recurrent adaptive problems.

Evolutionary psychology. Rant against marginalization of evolution in psychology academia.

Example – avoid mating with genetically similar organisms – deleterious recessives. But, care for genetic relatives. An identification problem. Computation of kinship index and adjustment of behavior based on it.

Implications for strong AI

- psychology map of humans so that it does what we mean

clues on how to achieve strong AI:

- dedicated intelligence
- pinpoint strategies
- “brilliant hacks”
- a mosaic of modules

Universe is too diverse to hold in our mind. Minds need to reduce.

15:20 – Tooby, Goertzel, Yudkowsky & Legg panel: Narrow and General Intelligence

…
Goertzel – AIXI is “fictitious”

Legg – AIXI is a useful model, and MC AIXI is useful in practice for limited problem sizes.
…

Goertzel – if evolution can harness large amounts of computation with a simple engine, then maybe we can achieve general intelligence with a small number of more general modules.

Legg – need domain specific algorithms to be efficient. But a general system can solve with unlimited power.

More discussion about how much domain specific algorithms you need. Possibly start with a limited set and develop more over time.

Legg – classes of algorithms that are not exactly domain specific, but applicable to multiple domain areas.

Singularity Summit 2010 – live blogging – day 1

miron — Sat, 14 Aug 2010 17:00:24 +0000

9:30 – Missed Michael Vassar‘s talk.

9:50 – Gregory Stock is talking. He is skeptical about progress in the bio realm. He says that the FDA is a damper on progress, but he also says that there are difficult problems. He brings up Alzheimer’s as an example. I think he is underestimating the power of info tech to change the way we do bio-science. Having read/write access to DNA, plus “in-silico” simulations will change the game.

Now he is talking about Silicon and saying that the complexity of computers rivals that of life. And now he is talking about the rapid exponential progress in DNA technology. As far as I understand, he is worried that we will create new life forms that will supersede humans. He is saying that human evolution is “not exponential”. I think he means that it’s a very slow exponential compared to tech.

He is talking about a “planetary super-organism” – merging of the biosphere with technological artifacts (cities, Internet). He says it’s beyond a metaphor. However, it’s not a replicator and does not have a boundary, so I am doubtful that it’s useful to think about what’s happening on earth in this way.

He is saying that the biosphere will be subsumed but still exist in the new merged organism. However, it seems to me that biology is not competitive because of the orders of magnitude in performance difference. He thinks that timescales will enlarge, but the characteristic frequencies of mechanical systems are much higher, so this doesn’t make sense to me.

He says we are in the midst of the Singularity.

Cyberspace – boundaries are weak (but what about encryption?), copies are easy (yes!). Uploading will lead to the disappearance/detachment of humans that take this route. Ways to stop evolution to a post-human future – singleton rule, super-organism communities. But it seems he thinks that stopping is not possible. People can’t affect because they are cells, and can’t affect the organism.

He says that people that follow regulation will not be the ones to make progress. The ones that do make progress are the nimble ones. Agreed.

11:00 – Ray Kurzweil – The Mind and how to build one

Singularity worry – is it feasible to emulate the brain.

We derived narrow AI capabilities from reverse engineering the auditor and visual systems. 5th computing paradigm – Moore’s law – planar silicon. 6th paradigm – 3d molecular computing.

He is complaining that critics set up straw-man arguments. More about exponential progress…

Why some people accept, and some people have emotional issues with accepting exponential change and it’s implications? Not because of sophistication, intelligence, etc. .

Quick recap of exponential progress.

Exponential progress in brain scanning. Resolution doubling every year. Showing a image of a simulation of a cortical column. Mentions blue brain project – Markram. Markram expects to reverse engineer the brain by 2018, Kurzweil thinks it will take until the 2020′s. Design of the brain is in the genome, which puts an upper bound on the code complexity strictly required – 25MB. ~10K code for cerebellum.

How to create a mind? Cortical columns are the basic modules. He equates these to LISP atoms. I think that’s too simplistic, because concepts are fluidly created. Nevertheless, I agree that these are the building blocks of the neocortex and it would be a good research direction for AI/AGI. It seems he is talking about de-novo mind design rather than uploading.

He says consciousness is a matter of faith and is not scientifically falsifiable. It is necessary to believe in it for morality. I’m not so sure – I agree with Dennet that it’s to do with having a self-model, which could be a scientific concept.

Wasn’t very satisfying as a roadmap to h+ AI.

Questions:

* Something about the collapse of the wave function related to consciousness. But the MWI doesn’t really have any wave function collapse, so I think that direction isn’t useful. Ray is saying the same thing.

* What level of detail for whole brain emulation? Ray says the point of WBE is to understand the principle of operation of the brain (i.e. reverse engineering). He says once we understand how cortical columns at a low level we can recode them at a higher level and have a much more computationally efficient simulation.

* Similar question – is molecular level simulation necessary? Ray responds that we don’t need to scan the whole brain at that resolution. Distinct types of neurons can be scanned with high resolution once. It seems he is again not talking about uploading.

12:00 – Ben Goertzel – AI for increased human healthspan

Human body as a machine.

The material in this talk is available on the h+ magazine site.

14:00 – Steven Mann: Humanistic Intelligence Augmentation and Mediation

Steven has glasses on which include a camera (I call this a speccam). The image from the camera is up on the left screen. He is concerned about surveillance. He is practicing sousveillance. He is drawing a diagram on a piece of paper instead of showing a slide. The diagram is visible through his speccam.

He is showing another speccam using his speccam. It re-renders the visible field on a 45 degree piece of glass, basically interposing a processing layer on vision – a heads up display.

A milling machine can replicate itself.

He talks about some patents he got on high dynamic range image stitching. Seems obvious – and kind of disappointing to see a software patent from him.

Cyborg-log. Continuous. He is talking about getting to a resolution below the perception granularity of a human. “Un-digital”.

He brought another person on stage that is talking about user-interfaces that keep humans in the loop as we get “un-digital”. Using high bandwidth interaction devices – tactile, visual. Musical instruments. They are demoing one on stage.

More about musical instruments and patents. This doesn’t seem very relevant. That’s about it for this talk.

14:50 – Mandayam Srinivasan: Enhancing our bodies and evolving our brains

Skipped this one.

15:25 – Brian Litt: The past, present and future of brain machine interfaces

Long review of existing tech – small number of electrodes, poor biocompatibility.

His new work – implant 2.5 micron thick flexible surfaces with 720 electrodes for capturing neuronal signals. Biocompatible, partly bioabsorbable (?).

Someone else’s work in rats. Memories played back during sleep at 6x original waking acquisition speed.

Speculates about self-assembling injectable electrodes.

16:15 – Demis Hassabis: Combining systems neuroscience and machine learning: a new approach to AGI

How to create meaning for symbols? CYC is brittle and cumbersome.

Search space of possible AGI solutions: regime 1 – small and dense search space. regime 2 – large sparse search space. In regime #2 it makes sense to use the brain as reference. He thinks regime #2 is the case. Arguments: evolution only produced it once, failure of AI field to date. I somewhat agree, except there might be a more dense area that we might be overlooking.

Abstract at one end of spectrum, whole brain emulation as other end. He thinks the abstract end is ad-hoc and unprincipled and WBE is 50 year in the future.

He advocates the middle way “systems neuroscience”. Extract algorithm from brain function. E.g. vision algorithm from the way vision areas work in the brain. Then you have a component for narrow AI or AGI. Another example is space perception (“grid cells”).

Reinforcement learning is another example. HNN is another.

This approach still seems somewhat ad-hoc to me. It seems likely that there are aspects of the way the brain works that cannot be identified as a separate module. These kind of aspects would be too intertwined with other modules and aspects to tease apart into a separate algorithm.

However, we might be able to achieve a different type of intellect with the modular reverse engineering approach that he advocates.

17:20 – Terry Sejnowski: Reverse-engineering brains is within reach

Should look at evolution when trying to understand the brain. A lot of conservation of genes even from “primitive” life forms.

Scale of the brain: 1e15 synapses, 1e11 neurons, 1e16 bits/s.

Temporal difference learning = dynamic programming. Dopamine system.

Use of TD for backgammon in 1994. Radio, radar, car and power grid applications in next 5 years.

Not very exciting.

17:50 – Dennis Bray: What Cells Can Do That Robots Can’t

The main thrust of this talk is that a lot of computation is done by cells. It seems to me that this computation is not relevant to the functioning of the mind. Most of it is keeping homeostasis in the cell and with the environment. When a person loses a limb, they maintain their mind and identity, so the cells in their limb could not have been “mission critical” to their identity and intelligence. Therefore the speaker’s argument seems irrelevant to the prospect of creating machine intelligence.

18:20 – Terry and Dennis are debating

Terry is saying that they will bite the bullet and do a “complete” molecular simulation. They did a monte-carlo simulation of a synapse over 100ms. Cool video of the simulation showing about 100 synapses – “MCELL”.

Dennis rebuts. He compares the brain to a neural network. He says that neurons are all different. I don’t see how that is related to the problem at hand – obviously you want to simulate the diversity of neurons present in a real brain by scanning one or by generating using embryo development simulation.

Terry: simulated a network of 300 generic neurons, and they learned useful things.

They seem to be “agreeing” and taking questions from the audience.

Dennis – there is nothing impossible in principle. He is now asking about details of Terry’s simulation. Terry – different proteins and molecules important at different times.

Dennis – there are so many different things to simulate. Terry – exponential progress.

I think they are mostly in agreement about only some molecular aspects being relevant and we need to simulate to find out. The main difference between them is probably related to intuition about exponential progress and the tools and resources it will make available in a short amount of time. Terry can see that.

Sir Martin Reese about the Future of the Cosmos

miron — Tue, 03 Aug 2010 18:27:53 +0000

Martin talks about the future of the cosmos and our responsibility to prevent existential risks at a long now foundation seminar.

Nice to see H+ memes coming from the president of the Royal Society.

H/T Tom McCabe @ Kurzweil AI

Long-Distance Wiring Diagram of the Monkey Brain

miron — Wed, 28 Jul 2010 06:41:30 +0000

Raghavendra Singh and Dharmendra S Modha published a paper in PNAS detailing 383 brain regions and 6,602 connections between them.