Ported Arcade game Space Invaders and other games onto home computer platforms. Developed new novel games.
Created a complete virtualized environment for the 8-bit Motorola MC6502 microprocessor for the DEC VAX and IBM 360/370 mainframe platforms.
Developed the World's first symbolic mathematical problem solver SymMath for the DOS platform, which covered fields from trigonometry, calculus, polynomials, matrix operations, differentiation and integrals. The entire system is based on Artificial Intelligence.
A hybrid fuzzy neural network based expert system.
A hybrid fuzzy symbolic/connectionist learning system. SC-Net was successfully applied to real-world domains such as: wafer fault diagnosis, segmentation of MRI images, determining credit-worthiness of applicants, military plane recognition using air-to-air radar return signals, network security and many other machine learning problems.
A new modified approach to the instance based learning theory. Instance based learning is augmented by neighborhood spheres and multi-pass training to improve both on generalization capabilities and storage requirements. Two models for creating neighborhood spheres are investigated and put in perspective with other IBL work. MPIL plays a pivotol role in the implementation of 124C41 vision system nearly twenty years later.
Created the World's first POC using SC-Net to demonstrate the inherent weakness of the LMHash code in network security.
Demonstrated the inherent limitations of PAC theory to be a reliable and useful means to estimate the size of training patterns required for machine learning problems. The findings of this basic research later formed the basis for creating ctSearch and 124C41. It also played a pivotal role in the development of 124C41's vision system, by allowing object recognizers to be constructed from as little as a few handful of examples.
TDL introduces a novel approach to pattern classification across multiple dimensions. It was also turned into a product at Universal Problem Solvers, Inc.
This research demonstrated for the first time that learning different problems iteratively and not independently of one another, can lead to significant speed-ups in training time, but also in terms of predictive capabilities. This research formed the basis for creating ctSearch and 124C41's learning ability.
EAS encompasses societies of similar agents, capable of cooperating and intelligently dividing complex tasks into manageable units. Autonomous, mobile agents capable of seeking each other out, based on their inherent problem solving qualities, and adept at forming loose societies.
A look at autonomous data mining agents and their applicability to automatically create expert networks by harvesting any type of data found online. Predicts the emergence of ctSearch and finally 124C41.
To investigate the ability of an EAS to survive in a real world ecology, a port of some of the earlier developed code was undertaken. Since the original code was written in Java a new port to Visual Basic was instituted to guarantee that EAS agents would only be able to thrive (i.e., move from one computer system to another, reproduce, communicate, devour nourishments (transformation from raw data sources into knowledge bases) on the Windows platform. In order to bypass all Windows security sentinels (i.e., anti-virus and firewall software) special capabilities had to be built into the EAS genome to guarantee survivability of the species in cyberspace. Visual Basic was found to be a good choice since it allowed to easily modify the bytecode of genome capability/need chunks on the fly thereby rendering software agents practically invisible to a systems defensive actors (signature detection based sentinels). Furthermore, the capability to detect whether EAS agents operated in a sandbox or on a real system became necessary to render behavioural-based sentinels obsolete. The EAS system in its end stage reached the point that it was no longer possible to guarantee how to shut it down, apart from turning off the power of an infected system and removing all hard drives and replacing them with new ones (to guarantee that rootkit-like/steganographic-ally hidden software instances would not survive). This test proved beyond a doubt the underlying capabilities of the EAS system and enabled the move to return development to a Java only version.
Developed complete frameworks based on NLP concepts and applied them to the partial automation of mobile Apps. One framework automated the translation into dozens of languages of the App menu and it's help sections.
Another framework introduced a novel detection for copyright infringement (i.e., copying and distributing multiple copies of a purchased App). This copy protection framework automatically inserted tracking code and markers into each App. Markers were unique to every App, allowing exact monitoring of duplicate instances and identifying from which purchased instance a copy originated. The framework was successfully used to track down thieves operating from China (down to the actual building) and who had been distributing illegally obtained Apps. Attempts to have the authorities prosecute these criminals - via lawyers in China - was not possible due to the exorbitant attorney fees, and as one attorney put it very slim chance of actually recovering any money by utilizing China's judicial system. Help from the US government (by contacting embassy staff) also yielded no results. In the end, it became necessary to resort to self-help by equipping future Apps with an automated disabling mechanism.
A third framework introduced a non-ip communications mechanism, which allowed all apps to communicate without presence of any dedicated device IP.
Finally, 3 Apps were created which used advanced NLP or natural language translation directly on the installed devices (no need to shuttle sensitive information to some cloud server and an extreme beneficiary solution to avoid all the personal data theft that occurred at Apple, Google, Microsoft, Amazon, Twitter, Facebook, Reddit and just about every other Silicon Valley based company years later). The Apps were:
All Apps could be combined with another App FindUs for indexing and search of information. Again, there was no need to transfer a single byte of information to any foreign servers. Everything was done on each device, guaranteeing maximum security of private data.
2016 - with Russia's invasion of Ukraine - all cooperation with Russian developers and connections to the existing customer base were severed. MobileTimes was dissolved and I continued - now full-time - with my earlier started sabbatical. The time was used to fully engage in basic research - primarily centring around human-level cognition. It was the birth hour of initially ctSearch (a context search engine) and later 124C41 (a deep inference engine). ctSearch's knowledge base grew by 2018 to about a trillion microfacts (a microfact is a basic unit of information with an associated certainty value - "water is wet (1.0)"). After 2018, ctSearch was relegated to a simple search bot, whereas 124C41 became the main backbone processing system - to which later all other developed bots would be linked (either directly or through other bots). At this point 124C41's knowledge bases had moved from static to dynamic, meaning that microfacts would be generated on the fly in real-time, no longer making it possible to estimate the amount of knowledge they contained. Since then EAS agents scour every nook and cranny on the Internet, Darknet and Intranets for data-sources to assimilate. These data-sources are turned into knowledge bases and automatically incorporated into the ecology of agents which have been built around 124C41. No human intervention is any longer required.
MobileTimes is also the world-wide leading company in the creation of web site and story polymorphism (i.e., generative AI). The company's products also achieved twice to pass the double-blind Turing test: 2007-2008 and 2014-2016. Some references to this work can be found here:
Generative AI 2012-2016, Generative AI 2017-2019, Generative AI 2020-EODAt first, a newly revised company of the original MobileTimes was created and appropriately named MobileTimesToday. Second, 124C41 received it's own company: Revelation Engine. In 2023 MobileTimesToday was split into two separate entities: a technology holding and a patent holding company. The later of which holds more than a dozen technological innovations and inventions.
I retired and started writing my memoirs.
Publication of the illustrated adult story book Tales from the Alternate Universe which was created largely by an immersive machine intelligence - 124C41 - and generative AI.
Feel Free To Drop Me A Line.
MobileTimesToday / Revelation Engine, New York, NY | San Jose, CA | Clearwater, FL