Large language modelA large language model (LLM) is a language model characterized by its large size. Their size is enabled by AI accelerators, which are able to process vast amounts of text data, mostly scraped from the Internet. The artificial neural networks which are built can contain from tens of millions and up to billions of weights and are (pre-)trained using self-supervised learning and semi-supervised learning. Transformer architecture contributed to faster training.
Neural machine translationNeural machine translation (NMT) is an approach to machine translation that uses an artificial neural network to predict the likelihood of a sequence of words, typically modeling entire sentences in a single integrated model. They require only a fraction of the memory needed by traditional statistical machine translation (SMT) models. Furthermore, unlike conventional translation systems, all parts of the neural translation model are trained jointly (end-to-end) to maximize the translation performance.
Transformer (machine learning model)A transformer is a deep learning architecture that relies on the parallel multi-head attention mechanism. The modern transformer was proposed in the 2017 paper titled 'Attention Is All You Need' by Ashish Vaswani et al., Google Brain team. It is notable for requiring less training time than previous recurrent neural architectures, such as long short-term memory (LSTM), and its later variation has been prevalently adopted for training large language models on large (language) datasets, such as the Wikipedia corpus and Common Crawl, by virtue of the parallelized processing of input sequence.
BehaviorBehavior (American English) or behaviour (British English) is the range of actions and mannerisms made by individuals, organisms, systems or artificial entities in some environment. These systems can include other systems or organisms as well as the inanimate physical environment. It is the computed response of the system or organism to various stimuli or inputs, whether internal or external, conscious or subconscious, overt or covert, and voluntary or involuntary.
Feature learningIn machine learning, feature learning or representation learning is a set of techniques that allows a system to automatically discover the representations needed for feature detection or classification from raw data. This replaces manual feature engineering and allows a machine to both learn the features and use them to perform a specific task. Feature learning is motivated by the fact that machine learning tasks such as classification often require input that is mathematically and computationally convenient to process.
Instructions per secondInstructions per second (IPS) is a measure of a computer's processor speed. For complex instruction set computers (CISCs), different instructions take different amounts of time, so the value measured depends on the instruction mix; even for comparing processors in the same family the IPS measurement can be problematic. Many reported IPS values have represented "peak" execution rates on artificial instruction sequences with few branches and no cache contention, whereas realistic workloads typically lead to significantly lower IPS values.
Generative pre-trained transformerGenerative pre-trained transformers (GPT) are a type of large language model (LLM) and a prominent framework for generative artificial intelligence. The first GPT was introduced in 2018 by OpenAI. GPT models are artificial neural networks that are based on the transformer architecture, pre-trained on large data sets of unlabelled text, and able to generate novel human-like content. As of 2023, most LLMs have these characteristics and are sometimes referred to broadly as GPTs.
Instruction set architectureIn computer science, an instruction set architecture (ISA), also called computer architecture, is an abstract model of a computer. A device that executes instructions described by that ISA, such as a central processing unit (CPU), is called an implementation. In general, an ISA defines the supported instructions, data types, registers, the hardware support for managing main memory, fundamental features (such as the memory consistency, addressing modes, virtual memory), and the input/output model of a family of implementations of the ISA.
Complex instruction set computerA complex instruction set computer (CISC ˈsɪsk) is a computer architecture in which single instructions can execute several low-level operations (such as a load from memory, an arithmetic operation, and a memory store) or are capable of multi-step operations or addressing modes within single instructions. The term was retroactively coined in contrast to reduced instruction set computer (RISC) and has therefore become something of an umbrella term for everything that is not RISC, where the typical differentiating characteristic is that most RISC designs use uniform instruction length for almost all instructions, and employ strictly separate load and store instructions.
Language modelA language model is a probabilistic model of a natural language that can generate probabilities of a series of words, based on text corpora in one or multiple languages it was trained on. Large language models, as their most advanced form, are a combination of feedforward neural networks and transformers. They have superseded recurrent neural network-based models, which had previously superseded the pure statistical models, such as word n-gram language model.
Scholarly peer reviewScholarly peer review or academic peer review (also known as refereeing) is the process of having a draft version of a researcher's methods and findings reviewed (usually anonymously) by experts (or "peers") in the same field. Peer review is widely used for helping the academic publisher (that is, the editor-in-chief, the editorial board or the program committee) decide whether the work should be accepted, considered acceptable with revisions, or rejected for official publication in an academic journal, a monograph or in the proceedings of an academic conference.
Behavior modificationBehavior modification is an early approach that used respondent and operant conditioning to change behavior. Based on methodological behaviorism, overt behavior was modified with consequences, including positive and negative reinforcement contingencies to increase desirable behavior, or administering positive and negative punishment and/or extinction to reduce problematic behavior. It also used Flooding desensitization to combat phobias.
Instructions per cycleIn computer architecture, instructions per cycle (IPC), commonly called instructions per clock is one aspect of a processor's performance: the average number of instructions executed for each clock cycle. It is the multiplicative inverse of cycles per instruction. While early generations of CPUs carried out all the steps to execute an instruction sequentially, modern CPUs can do many things in parallel.
Foundation modelsA foundation model (also called base model) is a large machine learning (ML) model trained on a vast quantity of data at scale (often by self-supervised learning or semi-supervised learning) such that it can be adapted to a wide range of downstream tasks. Foundation models have helped bring about a major transformation in how artificial intelligence (AI) systems are built, such as by powering prominent chatbots and other user-facing AI.
Instruction pipeliningIn computer engineering, instruction pipelining is a technique for implementing instruction-level parallelism within a single processor. Pipelining attempts to keep every part of the processor busy with some instruction by dividing incoming instructions into a series of sequential steps (the eponymous "pipeline") performed by different processor units with different parts of instructions processed in parallel. In a pipelined computer, instructions flow through the central processing unit (CPU) in stages.
Cycles per instructionIn computer architecture, cycles per instruction (aka clock cycles per instruction, clocks per instruction, or CPI) is one aspect of a processor's performance: the average number of clock cycles per instruction for a program or program fragment. It is the multiplicative inverse of instructions per cycle. The average of Cycles Per Instruction in a given process (CPI) is defined by the following weighted average: Where is the number of instructions for a given instruction type , is the clock-cycles for that instruction type and is the total instruction count.
CodeIn communications and information processing, code is a system of rules to convert information—such as a letter, word, sound, image, or gesture—into another form, sometimes shortened or secret, for communication through a communication channel or storage in a storage medium. An early example is an invention of language, which enabled a person, through speech, to communicate what they thought, saw, heard, or felt to others. But speech limits the range of communication to the distance a voice can carry and limits the audience to those present when the speech is uttered.
Applied behavior analysisApplied behavior analysis (ABA), also called behavioral engineering, is a psychological intervention that applies empirical approaches based upon the principles of respondent and operant conditioning to change behavior of social significance. It is the applied form of behavior analysis; the other two forms are radical behaviorism (or the philosophy of the science) and the experimental analysis of behavior (or basic experimental laboratory research).
Adaptive behaviorAdaptive behavior is behavior that enables a person (usually used in the context of children) to cope in their environment with greatest success and least conflict with others. This is a term used in the areas of psychology and special education. Adaptive behavior relates to everyday skills or tasks that the "average" person is able to complete, similar to the term life skills. Nonconstructive or disruptive social or personal behaviors can sometimes be used to achieve a constructive outcome.
ReproducibilityReproducibility, closely related to replicability and repeatability, is a major principle underpinning the scientific method. For the findings of a study to be reproducible means that results obtained by an experiment or an observational study or in a statistical analysis of a data set should be achieved again with a high degree of reliability when the study is replicated. There are different kinds of replication but typically replication studies involve different researchers using the same methodology.
